[Palaeontology, Vol. 53, Part 1, 2010, pp. 11–52]

THE AULACOSTEPHANIDAE (AMMONOIDEA) OF THE OXFORDIAN⁄ KIMMERIDGIAN BOUNDARY BEDS (UPPER JURASSIC) OF SOUTHERN ENGLAND by JOHN K. WRIGHT Department of Geology, Royal Holloway, University of London, Egham, Surrey, TW20 0EX, UK; e-mail [email protected]

Typescript received 30 August 2003; accepted in revised form 14 September 2009

Abstract: Four areas in southern England centred on the Late Oxfordian Pseudocordata Zone into Pseudoyo, Swindon and Westbury (Wiltshire), Bourton (north Dor- Pseudocordata and Evoluta Subzones is confirmed. In the set) and the Dorset coast near Weymouth (south Dorset) Early Kimmeridgian Baylei Zone, however, the evidence have yielded well-preserved late Oxfordian and early Kim- is that the sequence throughout much of southern England meridgian (Upper Jurassic) ammonites in abundance. is incomplete compared with more complete sequences These ammonites belong principally to the aulacostephanid such as that at Staffin in the Isle of Skye, with the pres- genera Ringsteadia and Pictonia, and their microconch ence of only one faunal biohorizon, the densicostata equivalents Microbiplices and Prorasenia. Systematic descrip- horizon. tions of these genera are included herein. Within the zonal and subzonal sequence of the English Oxfordian ⁄ Kimme- Key words: Oxfordian, Kimmeridgian, southern England, ridgian Stage boundary beds, the established subdivision of ammonites, biostratigraphy.

T he genera Ringsteadia and Pictonia, and their respective Arkell confined his attention solely to Microbiplices. Sykes microconch equivalents Microbiplices and Prorasenia, and Callomon (1979) subsequently described Upper comprise a major constituent of the late Oxfordian and Oxfordian Amoeboceras, but only one species of Ringste- early Kimmeridgian ammonite faunas of North-West adia, from the Isle of Skye. Matyja et al. (2006) figured Europe. They characterise the Sub-Boreal Province several Ringsteadia and Microbiplices from this locality, (Text-fig. 1), predominantly in England, northern France but, until the present work, there has been no compre- and western Germany and have been used extensively hensive modern study of these genera. Important work as zonal ⁄ subzonal indices throughout the province. on Ringsteadia in the doctoral thesis of Morris (1968) Their range extends northwards well into the Boreal remains unpublished. Province. Pictonia and Prorasenia have never been subject to South and southeastwards in the Sub-Mediterranean detailed modern study by British authors. Of the limited Province, most records of Ringsteadia in the literature work that has been performed, it is worth mentioning have proved to be of a separate, younger, Jura ⁄ Pomera- Buckman (1924b, 1927) and Spath (1935), who nian endemic group homeomorphic with Ringsteadia, but described respectively two new species of Pictonia and in fact having only a very ancestral connection with the two of Prorasenia. Significant faunas of Pictonia and North-West European faunas (Arkell 1956; Wierzbowski Prorasenia from the Isle of Skye were figured by Matyja 1970). et al. (2006). Numerous specimens of Pictonia from Perhaps surprisingly, the British aulacostephanid faunas northern France were figured by Tornquist (1896), but have attracted only a limited amount of research. Ringste- for many years this material lacked a satisfactory modern adia is comparatively well known because of the pioneer- account. This has now been rectified by Hantzpergue ing work of Salfeld (1917) who monographed the (1989). Recent stratigraphical research on the Kimmerid- Ringsteadia faunas of Wiltshire and Dorset. Buckman gian of Normandy has been summarized by Gallois (1919–30) figured several specimens of Ringsteadia. Arkell (2005). Specimens of Pictonia from Greenland have (1935, p. xxxii, 1940, p. lxv) promised a revision of Ring- been described by Birkelund and Callomon (1985) and steadia, but this task was later abandoned as ‘there are from Siberia by Mesezhnikov (1969). Sub-Mediterranean other tasks more urgent’ (Arkell 1947a, p. 352), and specimens recorded as Pictonia lack the regular deep

ª The Palaeontological Association doi: 10.1111/j.1475-4983.2009.00916.x 11 12 PALAEONTOLOGY, VOLUME 53

TEXT-FIG. 1. Extent of the Boreal, Sub-Boreal, Sub-Mediterranean and Mediterranean provinces across Europe, with the distribution of land and sea in the latest Oxfordian, and the principal continental localities mentioned in the text (after Ziegler 1982, Cope 1995 and Matyja and Wierzbowski 1995).

constrictions and flared ribs characteristic of Boreal and tinue to be useful in order to describe and define the Sub-Boreal forms and should be regarded as a separate range of variation existing within a biospecies, and, in endemic group (Pachypictonia) not directly connected. general, pre-existing specific names are retained here. Specimens which are taken to be the microconch coun- The retention of some of these names is necessary in terpart of Pictonia, are closely similar to the German particular in cases where formal subzones have been Prorasenia of Schindewolf (1925), and this name may named after morphospecies. No new specific names have also be used for the British forms. The aim of the pres- been introduced, however, and where morphotypes have ent paper is to provide the first full descriptions of been found whose morphologies are not covered by uppermost Oxfordian and lowermost Kimmeridgian existing morphospecies, these are described as variants aulacostephanids from southern England based on the of the closest morphospecies. latest understanding of the stratigraphy of these beds Groups of similar morphospecies have then been (Wright 2003). grouped into genera. In a very varied population, differ- ent generic names may have been applied to extreme members of one continuously variable population. An AULACOSTEPHANID TAXONOMY – attempt is made here to allocate all morphospecies of one ‘GENERA’, ‘SUBGENERA’ AND continuously variable biospecific taxon to one genus. The ‘SPECIES’ procedure of Arkell and others of using subgenera to group together assemblages of similar morphospecies is The extreme variability seen in aulacostephanids, partic- not adopted. Such groupings are best referred to by the ularly Ringsteadia, causes significant problems in taxon- name of their most typical morphospecies. When dividing omy. Early workers followed conventional classifications up a single evolving lineage into successive genera, this in terms of typological morphospecies and morphogen- can only be performed on grounds of convenience, per- era having stratigraphical ranges (often not specified or haps historically influenced. even known) – the ‘vertical’ classification. Such classifi- Dimorphism also presents taxonomic problems. Almost cations are purely subjective, arbitrary and hence unnat- all ammonite biospecies occur as dimorphs, the assump- ural. Classification today strives towards a natural tion being that the microconch forms were the males and taxonomy in terms of successions of ‘horizontal’ biospe- the macroconch forms the females (Makowski 1963). The cies (see e.g. Sylvester-Bradley (1956), or Callomon difference in adult size can be enormous – in the case of (1963, 1969, 1981, 1985) for ammonites in particular). Ringsteadia and its microconch counterpart Microbiplices, These are the phyletic transients of an evolving lineage, 30 mm (microconch) and 450 mm (macroconch). In the chronospecies or biospecies of some authors. Some some cases, authors have already decided to allocate one biospecies can be made up of half a dozen or more generic name for the combined dimorphic pair. However, morphospecies. However, morphospecific names con- the changes in characters used to define successive micro- WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 13 conch genera do not always occur at the same time as England and the northern part of southern England. In those used to define the successive macroconchs. This southern England from Westbury southwards, the latest being almost certainly the situation with Ringsteadia- Oxfordian is more arenaceous in facies, and the beds are Pictonia and Microbiplices-Prorasenia, separate morpho- allocated to the Sandsfoot Formation succeeded by the generic names for macroconchs and microconchs are Kimmeridge Clay Formation. retained here. In this southern area, sandy and ferruginous beds form convenient marker beds dividing up the clay succession (Text-fig. 3). The Sandsfoot Grit Member and associated AMMONITE STRATIGRAPHY Westbury Ironstone Member separate the Sandsfoot Clay and Ringstead Clay members, clays that are equivalent in This study describes ammonites collected from the Oxfor- age to parts of the Ampthill Clay succession of more dian ⁄ Kimmeridgian boundary beds principally in south- northern areas. The Osmington Mills Ironstone Member ern England. Localities in England referred to in the text marks the highest Sandsfoot Formation, succeeded by are marked in Text-figure 2B–D. The ammonites have the basal Kimeridge Clay Inconstans Bed, sandy, often been collected from predominantly clay facies rocks, phosphatic clays containing abundant, well-preserved much of sedimentation during the latest Oxfordian and ammonites. Kimmeridgian in England being argillaceous in nature. Within the clay members and formations, ammonites Beds are allocated to the Ampthill Clay Formation suc- are often crushed and poorly preserved, but at certain ceeded by the Kimmeridge Clay Formation in northern horizons, the development of micrites and ooidal iron- stones has resulted in the preservation of ammonites in three dimensions. The Marston Ironstone Bed of the Swindon area is a typical example (Wright 2003). A AB0° detailed study of the stratigraphy and sedimentology of these beds has been published by the author previously (Wright 2003). Logs of the sequences yielding ammonites 55° figured here are given in Text-figure 3.

Marston Ironstone. This is a 0.6 m bed composed of 5° limonite-ooid clay containing beautifully preserved micrite-filled ammonites present in the middle part of the Ampthill Clay Formation in the Wootton Bassett- South Swindon-South Marston-Shrivenham area of Wiltshire Ferriby (Text-fig. 3). Substantial new exposures of this bed were created during the construction of the Keypoint Industrial Distribution Site [the Keypoint site] at South Marston, east of Swindon, in 1999–2001. Highworth A complete list of the macrofauna was published by Swindon Oxford Shrivenham Wright (2003). Several of the best aulacostephanids are Wootton Bassett South London figured here in Plates 1–5. The following morphospecies Marston from this bed are described herein: Ringsteadia pseudocor- Westbury Bourton data (Blake and Hudleston), R. anglica Salfeld, R. brandesi Salfeld, R. pseudoyo Salfeld, R. sphenoidea Buckman, Dorset coast 0 50 km R. marstonensis Salfeld, R. bassettensis Spath and Micro- Weymouth sections 50° biplices cf. anglicus Arkell.

Fleet Lagoon Westbury Ironstone. This 4–5 m thick bed of ooidal iron- O.M.I OSMINGTON D WEYMOUTH exposures MILLS stone, included in the Sandsfoot Formation, is present in RINGSTEAD C the immediate vicinity of Westbury, Wiltshire (Talbot Sandsfoot Redcliff Black Head Ringstead sections Point Bran Bay Point Slipway 1974; Wright and Cox 2001) (Text-figs 2–3). WEYMOUTH BAY 0 1 2 3 km East Fleet section There is no opportunity of new ammonite collecting at TEXT-FIG. 2. Great Britain (A), England showing principal present, but many ammonites from Westbury are scattered localities mentioned in the text (B), Weymouth (south Dorset), through museum collections, and the following showing localities sampled (C) and Weymouth Bay (south aulacostephanid morphospecies are recognised herein: Dorset), showing localities sampled (D). Ringsteadia pseudoyo, R. pseudocordata, R. brandesi, 14 PALAEONTOLOGY, VOLUME 53

N South Ferriby South Marston-Swindon Westbury North Dorset South Dorset S Wootton Bassett (A 303)

Wyke Wyke Baylei Zone Siltstone Siltstone

Inconstans Bed Delta Clay Nana Bed Oxytoma Cementstone Nana Bed Nana Bed KIMMERIDGE CLAY FM KIMMERIDGE CLAY Inconstans Bed KIMMERIDGE CLAY FM KIMMERIDGE CLAY Inconstans Bed O.M.I. ? Evoluta Subzone RINGSTEAD ? ? CLAY RINGSTEAD CLAY 9 RED V 8 DOWN 7 WEST- IV 6 SAND SANDSFOOT Pseudocordata Subzone 5 -BURY 4 -IRON GRIT Marston Ironstone 3 -STONE III 2 ? 1 II SANDSFOOT FORMATION SANDSFOOT Pseudoyo Subzone I m 4 CORAL AMPTHILL CLAY FORMATION AMPTHILL CLAY 3 RAG SANDSFOOT 2 CLAY 1 FORMATION AMPTHILL CLAY 0

Coarse sand Mudstone Shells CORAL Fine to Calcareous RAG Limestone medium sand nodules Sideritic Silty clay Fe ooids nodules

TEXT-FIG. 3. Logs of the Oxfordian ⁄ Kimmeridgian boundary beds at South Ferriby, Swindon, Westbury and in Dorset (after Wright, 2003, fig. 2). I to V in south Dorset column subdivisions of the Sandsfoot Grit of Brookfield (1978) and Wright (1986). O.M.I., Osmington Mills Ironstone, passing laterally into the Ringstead Coral Bed. For location of sites see Text-figure 2.

R. anglica, R. marstonensis, R. bassettensis and Microbi- Although ammonites are frequently poorly preserved, plices anglicus. occasional specimens preserved in calcareous or phos- phatic concretions can be very well preserved. A list of Sandsfoot Grit. A series of sandstones, sandy clays and the macrofauna was published by Wright (1998). The sandy ooidal ironstones is present in the Upper Oxfordian following morphospecies, all from Unit III (Text-fig. 3), succession of the Sandsfoot Formation between Weymouth are described here: Ringsteadia brandesi, R. marstonensis and Bran Point, Dorset (Wright 1986, 1998; Text-fig. 3). and Microbiplices anglicus.

EXPLANATION OF PLATE 1 Fig. 1. Ringsteadia pseudoyo Salfeld, SM228, ·0.75. Fig. 2. Ringsteadia brandesi Salfeld, SM 229, ·0.6. Fig. 3. Ringsteadia pseudoyo Salfeld, SM192, ·0.8. Fig. 4. Ringsteadia marstonensis Salfeld, SM37, ·0.95. Fig. 5. Ringsteadia pseudocordata (Blake and Hudleston) var nudus nov., type of var., SM72, ·0.55. Figs 6–8. Microbiplices anglicus Arkell, Sandsfoot Grit, Pseudocordata Subzone, ·1.0; 6–7, Sandsfoot Castle, Weymouth, RHUL ic376a; 8, Black Head, DC61. Fig. 9. Prorasenia sp., Osmington Mills Ironstone, East Fleet section, NHM C75301, ·1.0. All specimens except 6–9 from the Marston Ironstone, Pseudoyo Subzone, Keypoint site, South Marston; 1–4 septate inner whorls; 9 is a largely complete microconch adult with the test preserved. Scale bars represent 10 mm. No scale bar = natural size. NB: On this and following plates, the end of the phragmocone is marked by an X. Where it is not clear in the figure, specimens that are wholly septate and portions of nonseptate body chambers are described as such in the plate descriptions; in some cases, the presence and extent of suture lines cannot be determined. PLATE 1

2

1

3

4

5

x

7

9 6 8

WRIGHT, Microbiplices, Prorasenia, Ringsteadia 16 PALAEONTOLOGY, VOLUME 53

Osmington Mills Ironstone. This sandy, micritic, limonite- the Oxford University Museum. A review of the collection ooid ironstone occupies the highest part of the Sandsfoot shows that the proportions of the morphospecies and Formation (Text-fig. 3), where it represents the top of the varieties present are almost exactly the same as in the Oxfordian succession between Weymouth and Ringstead Inconstans Bed of south Dorset. Bay (Brookfield 1978; Wright 2003; Williams 2003). It rests with a sharp, erosive break on the Ringstead Clay Others. During construction of the A303 Bourton Member. In the vicinities of Black Head and the Ring- bypass in north Dorset in 1991, tantalising glimpses of stead Bay slipway (Text-fig. 2D), the ironstone passes well-preserved Oxfordian ⁄ Kimmeridgian boundary beds laterally into coralliferous, micritic limestone named the faunas were seen (Bristow et al. 1991). Unfortunately, Ringstead Coral Bed (Wright 2003). the strata, of relatively deep water facies, were not con- A number of well-preserved Ringsteadia have been col- tinuously exposed in the road cutting, and the correla- lected by the author, and many more are to be found in tion of several small exposures was necessary to museum collections. The following morphospecies are produce a tentative section (Bristow et al. 1991, p. 141; described herein: R. frequens Salfeld, R. evoluta Salfeld, Text-fig. 3). R. pseudoyo, R. marstonense and Prorasenia sp. As was A Late Oxfordian ammonite fauna from the Ringstead noted by Wright (2003), almost all previous records of Clay Member, probably equivalent in age to the Osming- ammonites from the underlying Ringstead Clay have ton Mills Ironstone, comprises micrite-filled body cham- proved to be erroneous, the specimens in fact coming bers of Ringsteadia evoluta with occasional Ringsteadia from the basal part of the overlying Osmington Mills frequens, the inner whorls crushed flat except in occa- Ironstone, which is often argillaceous. sional calcareous concretions. An Early Kimmeridgian ammonite fauna comprising Inconstans Bed. Across much of southern England, the Pictonia sp. and Prorasenia cf. hardyi is contained within lowest 0.5–1.0 m of the Kimmeridge Clay Formation very fossiliferous concretions also containing well-pre- comprises a distinctive, shelly, sandy clay or argillaceous served bivalves and gastropods. This latter fauna was cor- sandstone containing phosphatised micrite-filled Pictonia related with that of the Inconstans Bed by Bristow et al. and Prorasenia as well as a prolific bivalve-brachiopod (1991). However, the Inconstans Bed, consisting of silty fauna (Brookfield 1978). This bed, which rests with an clay containing Pictonia sp. was exposed here in a tempo- unconformity on the underlying strata (Wright, 2003, fig. rary excavation at ST 7815 3050 in 2003. The concretions 8), was named by Arkell (1933, 1947b) the Inconstans yielding the Early Kimmeridgian fauna of Bristow et al. Bed after the common occurrence of the brachiopod (1991) occur below this level within extremely Torquirhynchia inconstans (J. Sowerby). fine-grained mudstone and may represent the earlier flod- Substantial numbers of ammonites from the Inconstans igarriensis horizon of Matyja et al. (2006). An excellently Bed are present in museum collections. The following preserved Rasenia cymodoce (d’Orbigny) was collected morphospecies are described herein: Pictonia densicostata from higher beds. All these specimens are in the BGS Buckman, P. seminudata (Buckman) and Prorasenia collection at Keyworth. hardyi Spath. Suggestions of flodigarriensis horizon faunas were also A large Inconstans Bed ammonite fauna was collected seen in the M40 excavations in Oxfordshire (Prorasenia from temporary exposures at Blagrove Farm and Wind- bowerbanki Spath and Amoeboceras praebauhini Salfeld, mill Hill, on the south-western outskirts of Swindon, by OUM Collection) and in the basal Kimmeridgian at J. H. Callomon in the 1990s. The specimens are now in South Ferriby Pit (P. bowerbanki, described below).

EXPLANATION OF PLATE 2 Fig. 1. Ringsteadia anglica Salfeld, trans. to R. pseudocordata, Keypoint site, South Marston, SM135, ·0.6; Fig. 2. Ringsteadia brandesi Salfeld, Keypoint site, South Marston; SM138, ·0.65. Fig. 3. Ringsteadia brandesi Salfeld, South Marston railway cutting, YORYM:2004.179, ·0.8. Figs 4–6. Ringsteadia brandesi, Sandsfoot Grit, Pseudocordata Subzone, Sandsfoot Castle, Weymouth, ·1.0; 4–5, RHUL ic105; 6, DC58. Figs 7–8. Microbiplices cf. anglicus. Keypoint site, South Marston, SM163, ·1.0. Fig. 9. Ringsteadia sphenoidea Buckman, Keypoint site, South Marston, SM96, ·0.75. All specimens except 4–6 from the Marston Ironstone, Pseudoyo Subzone; all septate inner whorls except 7–8. Scale bars represent 10 mm. No scale bar = natural size. PLATE 2

1 2

4 5 3

9

6

8

7

WRIGHT, Microbiplices, Ringsteadia 18 PALAEONTOLOGY, VOLUME 53

AMMONITE ZONES AND SUBZONES IN (Pseudocordata Subzone), as collections are limited (only THE SUB-BOREAL UPPERMOST 43 specimens seen). The proportion of morphotypes is OXFORDIAN – LOWERMOST similar to that of the Pseudoyo Subzone excepting for the KIMMERIDGIAN scarcity of R. pseudoyo in the Westbury fauna (pseudoyo group c. 2%, brandesi group c. 40%, marstonensis group c. The standard chronostratigraphical classification, based 60%). on Matyja et al. (2006), is shown in Text-figure 4. The ammonite fauna of the Evoluta Subzone (Osmington Mills Ironstone ⁄ Ringstead Coral Bed fauna) is quite distinctive (52 specimens). There is still UPPERMOST OXFORDIAN a substantial range in variation of shell shape, although not as marked as in the earlier subzones. R. pseudoyo Pseudocordata zone is still present (6%). Moderately involute forms are represented by the oval-whorled R. frequens (40%). The seven ‘species’ of Ringsteadia described by Salfeld Slim, evolute forms (R. evoluta, 52%) may be mark- (1917) were recorded indiscriminately throughout the edly constricted. One round-whorled form resembles Pseudocordata Zone, with the exception of R. evoluta at R. marstonensis. the top. Morris (1968) was able to demonstrate that in each of the three fossiliferous units, Marston Ironstone, Westbury Ironstone ⁄ Sandsfoot Grit and Osmington Mills LOWERMOST KIMMERIDGIAN Ironstone, one morphotype occurred more commonly than at the other horizons. These three faunas of Ringste- Baylei zone adia were characterised by R. pseudoyo, R. pseudocordata and R. evoluta, respectively. These subdivisions were listed Matyja et al. (2006) divided the Baylei Zone into two formally as subzones by Sykes and Callomon (1979) and subzones, a Densicostata Subzone overlain by a Norman- Wright (1980) and are listed as such in Text-figure 4, diana Subzone (Text-fig. 4). although they were subsequently used as faunal horizons A succession of four, possibly five faunal horizons of by Schweigert and Callomon (1997). Pictonia has been identified in the Sub-Boreal Province, The Pseudoyo Subzone, as typified by the fauna of combining evidence from localities between East Greenland the Marston Ironstone (197 specimens), is dominated (Birkelund and Callomon 1985), Staffin, Scotland (Morris by the evolute marstonensis group (51%). R. pseudocor- 1968, later revisions summarized in Morton and Hudson data and R. marstonensis are equally predominant, and 1995; Matyja et al. 2006), South Ferriby, Lincolnshire the coarsely ribbed R. bassettensis only amounts to 10% (Schweigert and Callomon 1997), Swindon and Dorset of the group. The involute brandesi group is also abun- (discussed herein) and Normandy (Hantzpergue 1989). dant (32%), and the highly involute pseudoyo group 1. P. flodigarriensis horizon: identified at Staffin and amounts to only 17% of the fauna. Prof. J. H. Callo- present in England in basinal areas such as north mon (pers. comm. 1998) has suggested that the low Dorset (Wessex Basin). Overstepped in shelf areas by proportion of the pseudoyo group here is because the the densicostata horizon. Marston Ironstone fauna lies near the top of the Pseu- 2. P. densicostata horizon: identified at Staffin and in doyo Subzone. England at South Ferriby via Oxford and Swindon to It is not possible to do a similar statistical analysis of the Dorset coast in the Inconstans Bed and missing the Westbury Ironstone ⁄ Sandsfoot Grit Ringsteadia fauna in a nonsequence in Normandy.

EXPLANATION OF PLATE 3 Fig. 1. Ringsteadia pseudocordata (Blake and Hudleston), Westbury Ironstone, Pseudocordata. Subzone, Westbury, SMC J44984, ·0.7. Fig. 2. Ringsteadia brandesi Salfeld, Marston Ironstone, Pseudoyo Subzone, Keypoint site, South Marston, SM19, ·0.6. Fig. 3. Ringsteadia pseudocordata (Blake and Hudleston), Marston Ironstone, Pseudoyo Subzone, Keypoint site, South Marston, SM114, ·0.65; Fig. 4. Ringsteadia marstonensis Salfeld, Sandsfoot Grit, Pseudocordata Subzone, Sandsfoot Castle, DC119, ·1.0. Fig. 5. Ringsteadia bassettensis Spath, Westbury Ironstone, Pseudocordata Subzone, Westbury, SMC J44972, ·0.85; Fig. 6. Ringsteadia marstonensis, Marston Ironstone, Pseudoyo Subzone, Keypoint Site, South Marston, SM29, ·1.0. All specimens septate inner whorls. Scale bars represent 10 mm. No scale bar = natural size. PLATE 3

1 2

3 4

6

5

WRIGHT, Ringsteadia 20 PALAEONTOLOGY, VOLUME 53

3. P. baylei horizon: identified at South Ferriby and (SMC); Natural History Museum, London (NHM); Devizes defined in Normandy in the lower limestone of the Museum (D); British Geological Survey, Keyworth (BGS); Fm. des Calcaires coquilliers (bed B in Hantzpergue Royal Holloway, University of London, comprising the former 1989, p. 54, fig. 12). collections of Chelsea, Bedford and Kings Colleges (RHUL), 4. ‘P. normandiana’ horizon: Normandy, upper lime- plus the author’s collection [DC, DK, SF and SM (now in BGS collection)]; University of Reading (UR), collection now stone of the Fm. des Calcaires coquilliers (bed D in housed at OUM; and the private collections of Professor J. H. Hantzpergue 1989). Some confusion attaches to the Callomon (JHC), Mr Clive Griffiths (donated to NHM) and naming of this horizon. Firstly, it is not certain that the late Mr Nick Samphier. One specimen loaned by York the lectotype of P. normandiana Tornquist, 1896, came Museum (YORYM). from this horizon, rather than from (3). Hantzpergue’s Several hundred ammonites have been examined in detail, claims to this effect must have been based on subse- and these are listed the Appendix. The collections of quently collected material and hence his interpretation J. H. Callomon (Baylei Zone, Swindon), BGS (Pseudocordata ⁄ of the species. If it did in fact come from (3), then Baylei Zones, A303) and N. Samphier (Pseudocordata ⁄ Baylei P. normandiana morphosp. would be merely a variant Zones, Ringstead) have only been assessed in general terms. of P. baylei biosp., as has at times been assumed in the Ammonite measurements: whorl height, umbilical width and past. Secondly, the name of the morphospecies was in whorl breadth are given conventionally as proportions of the shell diameter at which the measurements were made. fact regarded by Hantzpergue as a junior synonym of ‘P.’ thurmanni (Contejean, 1859), who therefore The following abbreviations are used: referred to horizon (4) as that of P. thurmanni. The D(phg), diameter at final septum; D(phg-ad), diameter of adult type and sole known specimen of this species came with uncoiling and ⁄ or crowding of final septae; D(phg-juv), from near Montbeliard (Doubs), eastern France, from juvenile; D(max), maximum diameter preserved; 0.x wh bch, beds immediately below a level with Pachypictonia fraction of body chamber preserved as a proportion of the indicatoria Schneidt (Amm. decipiens of Contejean: whorl; rpw, ribs per whorl. The principal illustrations of the specimens are in Plates 1– figured by Contini and Hantzpergue 1973). This 10. Drawings of cross-sections of species of Ringsteadia are strongly indicates the Cymodoce Zone. given in Text-figures 5 and 6, drawings of typical sutures of 5. ‘P. aff. normandiana’ horizon: East Greenland, Ringsteadia and Pictonia are given in Text-figure 7, and plots Birkelund and Callomon (1985) fauna 14, with of whorl height against umbilical diameter are given in Text- Amoeboceras bayi Birkelund and Callomon. A. bayi figure 8. occurs in the highest Baylei Zone at Staffin in Skye (Matyja et al. 2006). In morphospecific terms, the P. densicostata horizon as SYSTEMATIC PALAEONTOLOGY represented in the Inconstans Bed of south Dorset (89 specimens) is dominated by P. densicostata var. praebaylei nov. (47%), with P. densicostata (19%), P. densicostata Superfamily PERISPHINCTOIDEA Steinmann, 1890 var. dorsetensis nov. (10%), P. seminudata (14%) and Family AULACOSTEPHANIDAE Spath, 1924 P. seminudata var. ringsteadensis nov. (10%). As a preliminary to the descriptions below, a brief survey is presented of all existing nominal aulacostephanid REPOSITORY OF SPECIMENS AND genera that have been used for British Sub-Boreal CONVENTIONS Pseudocordata Zone and Baylei Zone material, but that are not used in the present work. A correlation of the The following ammonite collections have been studied: Oxford Sub-Boreal and Sub-Mediterranean zones and subzones University Museum (OUM); Sedgwick Museum, Cambridge referred to is given in Text-figure 4.

EXPLANATION OF PLATE 4 Figs 1–2. Ringsteadia bassettensis Spath var. strattonensis nov.; 1, specimen SM190, ·0.7; 2, specimen SM60, ·0.85. Each outer whorl is body chamber. Fig. 3. Ringsteadia marstonensis Salfeld, SM115, ·0.4. Figs 4–5. Ringsteadia bassettensis Spath; 4, specimen SM129, ·1.0; 5, specimen SM144, ·1.0. Fig. 6. Microbiplices anglicus Arkell, Sandsfoot Grit, Pseudocordata Subzone, Sandsfoot Grit, Sandsfoot Castle, ic377, ·1.0. All specimens except 6 from the Marston Ironstone, Pseudoyo Subzone, at the Keypoint site, South Marston. Scale bars represent 10 mm. No scale bar = natural size. PLATE 4

1 2

x

4

3

5

6

x WRIGHT, Microbiplices, Ringsteadia 22 PALAEONTOLOGY, VOLUME 53

Balticeras Dohm (1925, p. 34) Pachypictonia Schneid (1940, p. 89)

Type species. Balticeras pommerania Dohm, 1925 (pl. 5, figs 1– Type species. Pictonia indicatoria Schneid, 1940 (pl. 4, figs 3), from the Upper Jurassic of Zarnglaff, Pomerania (now 1–4), from the Malm Gamma 2 (Hypselocyclum Zone; Czarnogłowy, western Poland). Dohm inadvertently also used ?=Mutabilis Zone) of Zeegendorf in the Franconian Alb. Baltia in his text, but the intention clearly was to name the Schneid’s figures clearly show a coarsely ribbed macro- genus Balticeras. conch raseniid, with bold, prorsiradiate primary ribs on Dohm (1925) introduced the name for smooth-shelled, invo- the inner whorls. Pachypictonia is, in fact, a further mem- lute to moderately involute, discoidal ammonites collected from the incomplete, shelly, Oxfordian ⁄ Kimmeridgian succession of ber of the separate, Early Kimmeridgian Jura-Pomeranian shallow water platform carbonates in the Zarnglaff quarries. endemic group that includes also Pomerania and Baltic- There was little stratigraphical control over the collection of the eras. The use of Pachypictonia by Arkell (1948) for a coar- specimens, and their age, Late Oxfordian or Early Kimmerid- sely ribbed variant of Late Oxfordian Ringsteadia having gian, is unclear. Dohm (1925) included two quite separate forms almost smooth inner whorls was inappropriate and is not under the one name. B. ramlowi Dohm has a marked uncoiling adopted here. of the umbilical seam, with the outermost whorl only just in contact with the preceding whorl. The extremely involute B. pommerania Dohm has an unusual suture with a markedly Triozites Buckman (1924a, pl. 494) oblique second lateral lobe. Use of Balticeras as a subgenus for involute morphospecies of Ringsteadia, following Arkell et al. Type species. Triozites seminudatus Buckman, 1924a (pl. (1957), is not recommended. 494), in the distinctive matrix of the Inconstans Bed, Baylei Zone, of Ringstead Bay, Dorset. The affinities of Triozites are certainly with Pictonia. Pomerania Arkell (1937, p. 69) The suture is the same, with the lateral lobe slightly Type species. By original designation Pomerania dohmi Arkell, longer than the suspensive lobe, the development of a sec- 1937 (p. 69), nom. nov. for Pictonia baylei Dohm non Salfeld ond lateral and three small elements in the suspensive (Dohm, 1925, p. 32, pl. 5, fig. 6), from the Upper Jurassic of lobe. The presence of half a whorl of smooth phragmo- Zarnglaff, Pomerania (now Czarnogłowy, western Poland). The cone prior to the smooth body chamber is again very age of Pomerania is probably also Early Kimmeridgian. characteristic of Pictonia, as are the rectiradiate, rod-like The type species has 20–25 ribs per whorl from an early dia- primaries, although these occasionally may have a slight meter. It is certainly an aulacostephanid, but the prominent pri- ‘S’ curve, anticipating the rib style of Rasenia. However, mary ribbing suggests it is close to Rasenia. Pomerania was used no marked constrictions or flared ribs are visible on the in this sense for Early Kimmeridgian raseniids from the Polish inner whorls. T. seminudatus is thus used here, as Pictonia Holy Cross Mountains by Kutek (1968). By contrast, forms like seminudata, for nonconstricted varieties of Pictonia. P. schmidti (Dohm) are more densely ribbed on the inner whorls, with bold, coarse ribbing on the body chamber, which exceeds 500 mm in diameter. They may be perisphinctinids. Arkell (1937) included in his diagnosis fragmentary macro- Genus RINGSTEADIA Salfeld, 1913 conch perisphinctoid body chambers collected from the Upper Oxfordian Cautisnigrae Zone in Cambridgeshire. Arkell believed Type species. Ammonites pseudocordatus Blake and Hudleston, that the inner whorls of such specimens resembled Decipia. 1877 (p. 392, pl. xiii, fig. 1) by original designation of Salfeld Wright (1996) showed that the inner whorls of Arkell’s morpho- (1913, p. 427 and 1914, p. 129) from the Pseudocordata Subz- types were of the normal, biplicate, perisphinctoid type and one Westbury Ironstone of Westbury, Wiltshire. Holotype refig- named this taxon Pseudopomerania. Thus, Arkell’s (1947a) use ured by Buckman (1925a, pl. 560A, 560B). of Pomerania as a subgenus of Decipia is not followed here. Pseudocordata Zone forms, which Arkell assigned to Pomerania Remarks. Ringsteadia comprises an extremely variable are here regarded as coarsely ribbed, evolute varieties of Ring- group of large Late Oxfordian aulacostephanids (extending steadia.

EXPLANATION OF PLATE 5 Fig. 1. Ringsteadia anglica Salfeld, UR 10895. ·0.65, Westbury Ironstone, Pseudocordata Subzone, Westbury; Fig. 2. Ringsteadia pseudoyo Salfeld, Marston Ironstone, Pseudoyo Subzone, Keypoint site, Swindon, SM262, ·0.9; Fig. 3. Ringsteadia pseudoyo Salfeld, Osmington Mills Ironstone, Evoluta Subzone, Ringstead Bay, OUM J21186, ·0.85. Figs 4–5. Pictonia seminudata (Buckman), Inconstans Bed, Baylei Zone, Ringstead Bay: 4, OUM J67169, ·0.55; 5, OUM J13337, ·0.6. Specimens figured in 1–3 septate inner whorls; 5, outer whorl is body chamber. Scale bars represent 10 mm. PLATE 5

1

2

3

x 5 4

WRIGHT, Pictonia, Ringsteadia 24 PALAEONTOLOGY, VOLUME 53

Boreal Sub-Boreal Sub-Mediterranean zonesubzone zone subzone zone subzone

Mutabilis (pars) Hypselocyclum Lothari Modestum (pars)

Kitchini Cymodoce Guilherandense Platynota Subkitchini Desmoides Polygyratus Galar Normandiana Planula Baylei Planula KIMMERIDGIAN (pars)Bauhini STAGE Densicostata Hauffianum Bimammatum Bimammatum Evoluta Rosenkrantzi Hypselum Rosenkrantzi Pseudocordata Pseudocordata Marstonense Pseudoyo Grossouvrei Regulare Caledonica Bifurcatus Serratum Serratum Variocostatus Stenocycloides Koldeweyense Cautisnigrae OXFORDIAN (pars) OXFORDIAN Rotoides Glosense Cautisnigrae Transversarium Glosense IIovaiskii Nunningtonense (pars) Schilli

TEXT-FIG. 4. Zonal and subzonal scheme for part of the Upper Jurassic sequence in the Boreal and Sub-Boreal Provinces, and the correlation with that for the Sub-Mediterranean province, after Matyja and Wierzbowski (1997, fig. 4), Wright (2003, fig. 11) and Matyja et al. (2006). into the Early Kimmeridgian in the Sub-Mediterranean external and lateral lobes, a second lateral lobe, and a sepa- Province), maximum diameter up to 400 mm, shell shape rate, well-developed suspensive lobe with six distinctive ranging from strongly involute, discoidal to evolute, plan- elements. ulate. In the early whorls, the whorl section is invariably Salfeld (1917) recognised seven species of Ringsteadia, rounded, but between 30 and 50 mm diameter this providing detailed descriptions and illustrations. Buckman changes either to a flattened oval or to rounded-triangular. (1926) and Spath (1935) each added a further species to The whorl breadth ⁄ whorl height ratio thus varies from this list. These morphospecies have overlapping ranges 0.45 to 1.20. Some forms are smooth, but most have and will therefore co-occur to varying degrees at any one rounded primary ribbing fading gradually towards the end stratigraphical level. When they do, it has become clear of the phragmocone, with the secondaries arising irregu- that at such a level they represent merely the intergrading larly from the primaries often in sheaves of three or more. variants of a single biospecies. However, it is useful to Strong constrictions may be present, three or four per continue the use of morphospecific names in order to whorl, and in late forms these may be followed by a swol- facilitate description of the genus, for the range of varia- len simple rib. The variably complex suture comprises tion in morphology in Ringsteadia is so great.

EXPLANATION OF PLATE 6 Figs 1, 4–5. Ringsteadia frequens Salfeld, Osmington Mills Ironstone. Evoluta Subzone; 1, Black Head, JHC 933, ·1.0; 4, 5, Ringstead Bay; 4, SMC J44958, ·0.85; 5, SMC J44959, ·0.6. Fig. 2. Pictonia seminudata (Buckman), var. ringsteadensis nov. Inconstans Bed, Baylei Zone, Ringstead Bay, DK11, ·0.55. Figs 3, 6. Ringsteadia evoluta Salfeld, Osmington Mills Ironstone, Evoluta Subzone; 3, Black Head, DC139, ·1.0; 6, Ringstead Bay, DC125, ·1.0. Specimens figured in 1–4, 6 are septate inner whorls. Scale bars represent 10 mm. No scale bar = natural size. PLATE 6

1 2

3

4

5

x

6

WRIGHT, Pictonia, Ringsteadia 26 PALAEONTOLOGY, VOLUME 53

Attempts to subdivide Ringsteadia into subgenera (i.e. Type material. Two syntypes; lectotype (NHM 24089, Marston as in Arkell et al. 1957) bring nothing new to the taxo- Ironstone, Wootton Bassett), Salfeld, 1917, pl. 8, fig. 1a–b, desig- nomic problems and serve only to complicate those of nated by Buckman, 1926, pl. 672; paratype (NHM 36960, nomenclature. In place of a subgeneric classification, Marston Ironstone, South Marston). simple groupings based on typical morphospecies are preferred here. Text-figure 8A shows three major group- Measurements. Lectotype: D(max) 133 mm; at 133 mm, 0.47, 0.22, 0.17. BGS Zm2295: D(max) 109 mm; at 109 mm, 0.43, ings within the spread of involute to evolute forms in the 0.25, 0.22. OUM J14616 (figured Wright, 2003, fig. 9f): D(max) Pseudoyo Subzone. These can conveniently be referred to 114 mm; at 114 mm, 0.48, 0.26, 0.17, 25 ribs. OUM J14822: as three groups: (1), R. pseudoyo; (2), R. brandesi and (3), D(phg-ad) 175 mm; at 175 mm, 0.47, 0.23, 0.22; 0.1 wh. bc. R. marstonensis groups. Within each grouping, morpho- OUM J21186 (Pl. 5, fig. 3): D(max) 92 mm; at 92 mm; 0.49, species have been recognised by previous workers on the 0.29, 0.20. OUM J37527: D(max) 91 mm; at 91 mm, 0.46, 0.26, basis of differing whorl breadth or by differing strength of 0.22. SM192 (Pl. 1, fig. 3): D(phg-?ad) 147 mm; at 147 mm, ribbing. In the Pseudocordata Subzone (Text-fig. 8B), 0.49, 0.23, 0.15. SM228 (Pl. 1, fig. 1): D(max) 115 mm; at these three groups are still recognized, but with a prepon- 96 mm, 0.49, 0.26, 0.16. SM262 (Pl. 5, fig. 2): D(phg-?ad) derance of moderately evolute and evolute forms. In the 165 mm; at 102 mm, 0.46, 0.26, 0.19; at 155 mm, 0.47, 0.24, Evoluta Subzone (Text-fig. 8C), the spread from involute 0.24. to evolute forms is less, and there is less variation in strength of ribbing, so that only four morphospecific Description. This slim, discus-like form has an umbilical width varying from 15 per cent to 22.5 per cent of the diameter (Text- names have been used, R. pseudoyo, R. frequens, R. evoluta fig. 8A). On the inner whorls (Pl. 1, fig. 1; Pl. 5, fig. 2), the pri- and R. marstonensis. In what follows, the morphospecies mary ribbing is fine and prorsiradiate, with strong constrictions. of Ringsteadia in groups 1–3 are described in order of Above 20 mm, constrictions cease. The primaries are gently increasing evoluteness, followed by the descriptions of raised and slightly prorsiradiate, and broaden and pass into a R. frequens and R evoluta. sheaf of 3 or 4 secondaries carrying on the same prorsiradiate trend (Pl. 5, fig. 2). Above 90 mm, a smooth spiral band sepa- rates primaries and secondaries (Pl. 1, fig. 1). The secondaries 1. Ringsteadia pseudoyo group fade as they cross the venter. Some specimens are almost com- pletely smooth at 75 mm (Pl. 1, fig. 3). The whorl section is rounded ⁄ triangular, being broadest just above the umbilical Ringsteadia pseudoyo Salfeld, 1917 margin, and the sides sloping to meet the acutely rounded Plate 1, figures 1, 3; Plate 5, figures 2, 3; Plate 7, figure 4; venter (Text-fig. 5A–B). Text-figures 5A–B, E, 7B, I In specimens from the Ringstead Coral Bed (Pl. 5, fig. 3; Pl. 7, fig. 4), the umbilicus varies from 20 per cent to 23 per 1917 Ringsteadia pseudo-yo Salfeld, p. 74, pl. 8, cent of the diameter. Moderately strong, well-spaced primaries figs 1a–b, 2a–b. and secondaries are developed at 35 mm (Pl. 7, fig. 4). Constric- ?1978 Ringsteadia sp.; Rozak and Brochwicz-Lewinski, tions interrupt the rib pattern, but frequently sheaves of three pl. 4, fig. 3a–b. secondaries arise from primaries. The primary ribs fade earlier 2003 Ringsteadia pseudoyo Salfeld; Wright, fig. 9f. than the strongly developed secondaries. There is a subtle angle 2006 Ringsteadia ex. gr. pseudoyo Salfeld; Matyja et al., between the slightly prorsiradiate primaries and rectiradiate fig 4b.

EXPLANATION OF PLATE 7 Fig. 1. Ringsteadia frequens Salfeld, Osmington Mills Ironstone, Evoluta Subzone, ?Ringstead Bay, OUM J13104, ·0.6. Fig. 2. Pictonia densicostata Buckman, var. praebaylei nov., type of var., Inconstans Bed, Baylei Zone, Ringstead Bay, SMC J47870, ·0.5. Fig. 3. Ringsteadia evoluta Salfeld, largely complete juvenile macroconch, Osmington Mills Ironstone, Evoluta Subzone, Osmington Mills, JHC 936, ·1.0. Fig. 4. Ringsteadia pseudoyo Salfeld, Osmington Mills Ironstone, Evoluta Subzone, Ringstead Bay, OUM J37757, ·0.85. Figs 5–6. Pictonia densicostata Buckman, var. dorsetensis nov. Inconstans Bed, Baylei Zone, Ringstead Bay, DK6, ·1.0. Fig. 7. Ringsteadia frequens Salfeld, Osmington Mills Ironstone, Evoluta Subzone, Ringstead Bay, DC141, ·0.75. Fig. 8. Prorasenia sp.,, Osmington Mills Ironstone, Evoluta Subzone, East Fleet section, NHM C75302, ·1.0. Fig. 9. Prorasenia bowerbanki Spath, Kimmeridge Clay, Oxytoma Cementstone, Baylei Zone, South Ferriby Clay Pit, SF1, ·1.0. Specimens figured in 4–6 are septate inner whorls; 8 is a partially complete, microconch adult with the test preserved, showing uncoiling of the umbilical seam; 9 is a complete, microconch adult with lappets, and much of the test preserved. Scale bars = 10 mm. No scale bar = natural size. PLATE 7 x

2

1 x

3

4

x 5

8

7

6

9

WRIGHT, Pictonia, Prorasenia, Ringsteadia 28 PALAEONTOLOGY, VOLUME 53 secondaries. These then have a distinct forward swing at 2. Ringsteadia brandesi group the ventro-lateral edge, giving a slight S-shaped curve to the ribbing. Ringsteadia brandesi Salfeld, 1917 Plate 1, figure 2; Plate 2, figures 2–6; Plate 3, figure 2; Text- Stratigraphical range. These markedly involute forms of Ringste- figures 5D, G–H, 7C–F, J adia are moderately common in the Marston Ironstone (Pseu- doyo Subzone) of South Marston. R. pseudoyo is rare in the 1917 Ringsteadia brandesi Salfeld, p. 77, pl. 9, fig. 1a–b, Pseudocordata Subzone and also occurs sporadically in the south pl. 10, fig. 3a–b. Dorset Evoluta Subzone. non 1917 Ringsteadia brandesi Salfeld, pl. 11, fig. 3a–b. ?1917 Ringsteadia brandesi Salfeld, p. 78. Remarks. The Evoluta Subzone forms of R. pseudoyo are 2007 Ringsteadia cf. anglica Buckman; Rogov and not quite as involute as those in the Pseudoyo Subzone, Kiselev, pl. 4, figs 1–2. with a whorl section having the sides more nearly parallel than in the early R. pseudoyo. A specimen close to R. pseu- Type material. Two syntypes from the Marston Ironstone; lecto- doyo, involute, with early fading of the primaries and dense type now designated; Salfeld, 1917, pl. 9, fig. 1a–b, NHM 50770, secondaries, was figured from the lower Hypselum Subz- from South Marston; paratype NHM 15425 from Wootton one (mid-Pseudocordata Zone) of Czestochowa in Poland Bassett. by Rozak and Brochwicz-Lewinski (1978). Measurements. Lectotype: D(max) 138 mm; at 138 mm, 0.42, 0.26, 0.26. BGS GSM7717: D(max) 138 mm; at 138 mm, 0.42, 0.21, 0.23. BGS HBW2644: D(phg-ad) 258 mm, D(max) Ringsteadia sphenoidea Buckman, 1926 275 mm, at 167 mm, 0.42, 0.23, 0.29. SM19 (Pl. 3, fig. 2): Plate 2, figure 9; Text-figure 5C D(max) 145 mm; at 145 mm, 0.42, 0.27, 0.29. SM138 (Pl. 2, fig. 2): D(max) c. 160 mm; at 82 mm, 0.42, 0.30, 0.31, 19 ribs. 1926 Ringsteadia sphenoidea Buckman, pl. 672. SM229 (Pl. 1, fig. 2): (D(phg-?ad) 151 mm; at 149 mm, 0.41, 0.24, 0.26. YORYM:2004.179 (Pl. 2, fig. 3): D(max) 99 mm; at Type material. Holotype (NHM C41691) from the Marston 99 mm, 0.43, 0.31, 0.26. Ironstone of Wootton Bassett. Description. R. brandesi has the umbilical width varying from 23 Measurements. Holotype: D(max) 180 mm; at 170 mm, 0.48, per cent to 31 per cent of the total diameter. The early primary 0.27, 0.16. SM96 (Pl. 2, fig. 9): D(max) 165 mm; at 162 mm, ribs are broadly rounded and slightly prorsiradiate. On coarser 0.48, 0.30, 0.18. ribbed forms (Pl. 2, fig. 2), the secondaries bifurcate or trifurcate just below the mid-point of the whorl side, continuing along the Description. Comparison of Plate 2, figure 9 with the specimen line of the primaries and curving forwards slightly over the ven- figured as R. pseudoyo in Plate 1, figure 3 shows that R. sphenoi- ter. On finer ribbed forms (Pl. 2, fig. 3; Pl. 3, fig. 2), the prima- dea is just as involute as R. pseudoyo but is distinguished by its ries pass indistinctly into sheaves of three secondaries. At much broader, triangular whorl section (Text-fig. 5C). The spec- 70 mm diameter, the primary ribs begin to fade, although the imen figured matches Buckman’s holotype perfectly, and this is secondaries continue more prominently. Above 110 mm, almost a distinctive and easily recognised morphospecies. all forms are smooth, although one specimen has well-developed ribbing at 180 mm. At 40 mm diameter, the whorl breadth is Stratigraphical range. R. sphenoidea is a moderately common equal to the whorl height, but thereafter a slim whorl section form in the Pseudoyo Subzone of South Marston. It is not develops, with the maximum breadth being at a quarter or a known in the Pseudocordata or Evoluta subzones. third of the whorl height (Text-fig. 5D, G–H).

EXPLANATION OF PLATE 8 Fig. 1. Ringsteadia frequens Salfeld, Osmington Mills Ironstone, Osmington Mills, Evoluta Subzone, NHM C93796, ·0.85. Fig. 2. Pictonia seminudata (Buckman), Inconstans Bed, Ringstead Bay, OUM J29499, ·1.0. Figs 3–5. Ringsteadia evoluta Salfeld, Osmington Mills Ironstone, Evoluta Subzone; 3, 4, Black Head; 3, JHC 931, ·0.7; 4, JHC 932, ·0.8; 5, Osmington Mills, DC122, ·0.75. Fig. 6. Pictonia densicostata Buckman, largely complete juvenile macroconch, Inconstans Bed, Baylei Zone, Ringstead Bay, DK9, ·1.0. Figs 7–8. Prorasenia cf. hardyi Spath, Inconstans Bed, Baylei Zone, Ringstead Bay, ·1.0: 7, DK10; 8, DK34. Specimens figured in 1, 3 and 7 are septate inner whorls; 2 is a nearly complete juvenile macroconch. Scale bars represent 10 mm. No scale bar = natural size. PLATE 8

1 2

x

4

3

5 6

7

8

x

x WRIGHT, Pictonia, Prorasenia, Ringsteadia 30 PALAEONTOLOGY, VOLUME 53

Stratigraphical range. R. brandesi is one of the most common NHM C50768: D(max) 135 mm; at 105 mm, 0.43, 0.28, 0.29. morphospecies in the Pseudoyo Subzone. It is less frequent in the SM135 (Pl. 2, fig. 1): D(max) 157 mm; at 129 mm, 0.41, 0.20, Pseudocordata Subzone, and in the Evoluta Subzone its place as a 0.27. UR 10895 (Pl. 5, fig. 1): D(max) 148 mm; at 148 mm, moderately involute species of Ringsteadia is taken by R. frequens. 0.40, 0.20, 0.28.

Remarks. The name R. brandesi has previously been Description. R. anglica is a slim, discoidal, moderately evolute applied to Evoluta Subzone Ringsteadia. However, careful form, with the umbilical margin scarcely raised at all (Text- study of the lectotype and paratype shows that both came fig. 5F). At small diameters, the ribbing is quite bold (Pl. 5, fig. from the Marston Ironstone, and are thus from the Pseu- 1), with well-differentiated primaries and secondaries, and fre- doyo Subzone. A third specimen described as R. brandesi quent-marked constrictions. On the middle whorls, the primary ribbing may be quite distinct (25 ribs per whorl at 150 mm, came from the Osmington Mills Ironstone (Salfeld, 1917, Pl. 2, fig. 1), or it may be faint or even non-existent. The coiling pl. 11, fig. 3a, b) and is regarded here as R. frequens. The at these diameters is quite intense, with considerable overlap of differences between R. brandesi and R. frequens are dis- the previous whorl and the umbilical diameter less than the cussed under R. frequens below. Three specimens of whorl height. At 120 mm, uncoiling starts. At large diameters, R. brandesi are recorded from the Upper Korallenoolith the umbilical diameter is much greater than the whorl height. of the Hannover region (Galgenberg, Salzhemmendorf) by Salfeld (1917). A specimen very close to R. brandesi, Stratigraphical range. R. anglica typifies the Pseudocordata Subz- from Trzebinia, southern Poland, and described originally one. It is much less common in the Pseudoyo Subzone and has as Proplanulites teisseyrei Siemiradzki (Siemiradzki 1891), never been recorded from the Evoluta Subzone. is in the Siemiradzki Collection in Krakow (Głowniak and Wierzbowski 2007). One specimen of R. brandesi was Remarks. Both Buckman (1921, pl. 225) and Arkell figured by Rogov and Kiselev (2007) from the Upper Ox- (1933, pl. 38) stated that the specimens from the West- fordian of Mikhalenino, Unzha River Basin, North-east of bury Ironstone which they were figuring as R. anglica Moscow. were ‘topotypes’, i.e. from the same locality as the type, but neither author formally designated the sole specimen from Westbury figured by Salfeld in his type series of Ringsteadia anglica Salfeld, 1917 R. anglica as lectotype. The species is distinguished from Plate 2, figure 1; Plate 5, figure 1; Text-figure 5F other moderately evolute morphospecies of Ringsteadia by the extremely slim whorl section, with maximum breadth 1917 Ringsteadia anglica Salfeld, p. 76, pl. 8, fig. 3a–b, pl. 10, fig. 2a–b, pl. 13, fig. 1. of the whorl being at the mid-point of the whorl side 1921 Ringsteadia anglica Salfeld; Buckman, pl. 225. (Text-fig. 5F). non 1933 Ringsteadia anglica Salfeld; Arkell, pl. 38, fig. 1. non 2007 Ringsteadia cf. anglica Buckman; Rogov and Kiselev, pl. 4, figs 1–2. 3. Ringsteadia marstonensis group

Type material. Three syntypes; lectotype now formally desig- Ringsteadia pseudocordata (Blake and Hudleston, 1877) nated (Salfeld 1917, pl. 13, fig. 1, BGS GSM25502, from the Plate 1, figure 5; Plate 3, figures 1, 3; Text-figures 5I–J, 7A Westbury Ironstone); paratypes NHM C50768 from the Ampt- hill Clay (?Marston Ironstone) of Shrivenham and NHM C322 1877 Ammonites pseudocordatus Blake and Hudleston, (‘Wiltshire’). p. 392, pl. 13, fig. 1. non 1917 Ringsteadia pseudocordata (Blake and Hudleston); Measurements. BGS GSM97551: D(max) 172 mm; at 172 mm, Salfeld, pl. 10, fig. 1a–b. 0.30, 0.20, 0.44, 32 ribs.

EXPLANATION OF PLATE 9 Figs 1, 3, 7. Pictonia densicostata Buckman, var. praebaylei nov. 1, SMC J47913, ·0.9; 3, NHM C93791, ·0.85. 7, NHM C93793, ·0.45. Fig. 2. Pictonia densicostata Buckman, SMC J47867, ·0.85. Fig. 4. Prorasenia bowerbanki Spath, Kimmeridge Clay, Baylei Zone, Bourton Bypass, DK35, ·1.0. Figs 5–6. Pictonia densicostata Buckman, var. dorsetensis nov., type of var., DK31; 5, ·0.65; 6, ·0.9. Fig. 8. Pictonia densicostata Buckman var. grandicostata nov., DK32, ·0.6. Specimens figured in 1, 2, 3, 7, 8 from the Inconstans Bed, Baylei Zone, of Ringstead Bay; 5–6 from the Inconstans Bed of Black Head; 1–2 and 5–6 are septate inner whorls; 4 has a fragment of the body chamber showing a terminal constriction and part of a lappet. Scale bars represent 10 mm. No scale bar = natural size. PLATE 9

2 1

3

5

4

6

7

x x 8

WRIGHT, Pictonia, Prorasenia 32 PALAEONTOLOGY, VOLUME 53

non 1925 Ringsteadia pseudo-cordata Blake and Hudleston Stratigraphical range. The Pseudoyo and Pseudocordata Subz- emend Salfeld; Dohm, p. 27, pl. 1, figs 8, 9, pl. 3, ones are characterized by this moderately evolute form of Ring- fig. 3, pl. 6, figs 9, 10, pl. 8, fig. 2, pl. 9, figs 1, 3. steadia, which, next to R. brandesi, is the most common 1925a Ringsteadia pseudocordata (Blake and Hudleston); Ringsteadia in the two lower subzones of the Pseudocordata Buckman, pls. 560A, 560B. Zone. 1925b Ringsteadia frequens Salfeld; Buckman, pl. 589. 1933 Ringsteadia anglica Salfeld; Arkell, pl. 38, fig. 1. Remarks. R. pseudocordata is characterized by sharp, regu- 2003 Ringsteadia pseudocordata (Blake and Hudleston); lar ribbing on the inner whorls (Pl. 3, fig. 1), whereas the Wright, fig. 9a. similarly evolute R. evoluta frequently has irregularly ribbed inner whorls much affected by constrictions (Pl. 8, Type material. Holotype (BGS GSM46264) from the Westbury figs 3, 5). The specimens from north-west Poland figured Ironstone of Westbury, Wiltshire, figured by Buckman (1925a, as R. pseudocordata by Dohm (1925) show a combination pls 560A, 560B). Var. nudus; type of var. (SM72) from the Mar- ston Ironstone of the Keypoint site, Swindon. of features not found in English specimens and should not be placed in this species. The ribbing of the inner Measurements. Holotype: D(max) 190 mm; at 190 mm, 0.36, whorls is more dense than that of the specimens 0.25, 0.35. BGS GSM97554: D(phg-?ad) 230 mm, D(max) described here, with a smooth outer phragmocone and a 290 mm; at 290 mm, 0.35, 0.16, 0.36, 27 ribs at 240 mm; 0.3 wh body chamber with coarse, rounded ribbing. The cross- bch. OUM J14615 (figured Wright, 2003): D(max) 119 mm; at section is flat sided on the middle whorls, then develop- 119 mm, 0.38, 0.30, 0.34, 32 ribs. ing a ventrolateral spiral depression to give an acute, SMC J44984 (pl. 3, fig. 1): D(max) 116 mm; at 116 mm, 0.36, rounded venter on the body chamber. However, a Polish 0.28, 0.35, 29 ribs at 97 mm. specimen very close to R. pseudocordata, described origi- Var. nudus: SM72. Type of var. (Pl. 1, fig. 5), SM72: D(phg-ad) nally as Proplanulites teisseyrei (Siemiradzki 1891), is pres- 150 mm, D(max) 190 mm, 0.5 wh bch; at 190 mm, 0.35, 0.22, ent in the Siemiradzki Collection in Krakow, from 0.34. SM244: D(phg-ad) 155 mm, D(max) 190 mm, 0.5 wh bch. Trzebinia, southern Poland (Głowniak and Wierzbowski 2007). Description. R. pseudocordata is the most involute of the R. mar- stonensis group, with the umbilical width is about equal to the whorl height (31–36% of the diameter). The development of rib- Ringsteadia marstonensis Salfeld, 1917 bing is very variable. Strong, wiry, slightly prorsiradiate primary ribs, 27 per whorl, are present at 20 mm, with strong constric- Plate 1, figure 4; Plate 3, figures 4, 6; Plate 4, figure 3; tions (Pl. 3, fig. 1). At 40–45 mm, the primary ribs become flat- Text-figures 6A, C, 7G–H ter and more rounded. The secondaries, first seen at 75 mm in Plate 3, figure 1, are quite weak, 2 or 3 per primary. They usu- 1917 Ringsteadia marstonensis Salfeld, p. 83, pl. 11, ally arise in sheaves of three secondaries, although there is some figs 1a–b. bifurcation. The whorl thickness is at its maximum just above non 1925 Ringsteadia marstonensis Salfeld; Dohm, p. 28, the umbilical margin, and the sides then slope to the rounded pl. 2, figs 5–6, pl. 3, fig. 1 venter, giving an elliptical whorl section (Text-fig. 5I–J). There is non 1925 Ringsteadia cf. marstonensis Salfeld; Dohm, every variation from strongly ribbed forms through more gently p. 28, pl. 6, figs 7, 11. ribbed ones (Pl. 3, fig. 3) to forms that are smooth even at non 1962 Ringsteadia marstonensis Salfeld; Wilczynski 80 mm (Pl. 1, fig. 5). Such weakly ribbed or smooth forms form pp. 64-66, pls 1, 2. a distinct majority of the specimens collected at South Marston non 1962 Ringsteadia cf. marstonensis Salfeld; Wilczynski, and may be distinguished as var. nudus nov. They become adult p. 66, pl. 3, fig. 1 at much earlier diameters than the ribbed forms (SM72, SM244 non 1988 Ringsteadia marstonensis Salfeld; Mesezhnikov, above). If ribbing is present, it is maintained to at least 190 mm pl. 9, fig. 13. on the phragmocone, as seen on the holotype.

EXPLANATION OF PLATE 10 Figs 1, 3. Pictonia seminudata (Buckman) var. ringsteadensis nov.: 1, type of var., DK8, · 0.7; 3, OUM J37712, · 0.5. Fig. 2. Pictonia densicostata Buckman var grandicostata nov., type of var, SMC J47868, · 0.75. Fig. 4. Pictonia densicostata Buckman, OUM J67110, · 0.85. Figs 5–6. Prorasenia cf. hardyi Spath, · 1.0: 5, SMC J47351; 6, DK33. All specimens from the Inconstans Bed, Baylei Zone, of Ringstead Bay. In 3, the outer whorl as preserved is body chamber; 5 is largely complete, with the aperture including a prominent lappet partly preserved on the reverse behind the label; 6 is a body chamber fragment. Scale bar represent 10 mm. No scale bar = natural size. PLATE 10

1

2

x

x

3

4

5 6

x

WRIGHT, Pictonia, Prorasenia 34 PALAEONTOLOGY, VOLUME 53

TEXT-FIG. 5. Whorl sections of Ringsteadia from South Marston. A C B D SM192 SM96 SM228 SM138 R. pseudoyo R. sphenoidea R. pseudoyo R. brandesi Pl. 1, fig. 3 Pl. 2, fig. 9 Pl. 1, fig. 1 Pl. 2, fig. 2

E SM262

R. pseudoyo G Pl. 5, fig. 2 F SM229 SM135 R. brandesi R. anglica Pl. 1, fig. 2 Pl. 2, fig. 1

J H I SM114 SM19 SM72 R. pseudocordata R. brandesi R. pseudocordata Pl. 3, fig. 3 Pl. 3, fig. 2 Pl. 1, fig. 5

Type material. Salfeld (1917) figured two syntypes, but one was trifurcate into less prominent, rectiradiate secondaries, this bold subsequently designated the holotype of R. bassettensis by Spath and rod-like ribbing continuing to at least 210 mm (Pl. 4, (1935). The remaining syntype (Salfeld, 1917, pl. 11, figs 1a, b, fig. 3). The secondaries arise in a very irregular manner (Pl. 1, NHM C15427), from the Marston Ironstone of South Marston, fig. 4). Bifurcation may take place low on the whorl side or near thus becomes the holotype. the ventro-lateral margin. Primary ribs may pass into a sheaf of 3 or 4 secondaries or may continue over the whorl side as a sim- Measurements. Holotype: D(max) 148 mm; at 148 mm, 0.36, ple rib (Pl. 4, fig. 3). The secondaries then run strongly over the 0.32, 0.38. BGS GSM97555: D(max) 142 mm; at 142 mm, 0.30, venter. The holotype has sheaves of secondaries, 6 per primary, 0.19, 0.44. DC126: D(max) 210 mm; at 145 mm, 0.39, 0.32, at 150 mm. The ribbing may fade completely, Plate 3, figure 6 0.37. SM115 (Pl. 4, fig. 3): D(phg-?ad) 205 mm; at 201 mm: showing the shell almost smooth at 120 mm. 0.35, 0.25, 0.42; 17 ribs per half whorl at 180 mm. Stratigraphical range. R. marstonensis is common in the Pseu- Description. R. marstonensis has the umbilical width ranging doyo and Pseudocordata subzones, and scarce in the Evoluta from 36 per cent to 42 per cent of the shell diameter, as against Subzone, one specimen from the Osmington Mills Ironstone whorl heights ranging from 34 per cent to 37 per cent (‘e’, Text- (DC126) showing the characteristic features very well. fig. 8A). The whorl breadth is about two thirds of the whorl height (Text-fig. 6A, C). On the early whorls, at 30 mm diameter Remarks. R. marstonensis was introduced by Salfeld (Pl. 3, fig. 4), bold, prorsiradiate primaries bifurcate or occasionally (1917) for all evolute, round-whorled Ringsteadia. The WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 35

TEXT-FIG. 6. Whorl sections of Ringsteadia from South Marston and from south Dorset. A B SM115 SMC J44972 D SM190 R. marstonensis R. bassettensis Pl. 4, fig. 3 Pl. 3, fig. 5 R. bassettensis var. strattonensis Pl. 4, fig. 1

C SM37

R. marstonensis Pl. 1, fig. 4

G SM60 E F SM129 SM144 R. bassettensis var. strattonensis R. bassettensis R. bassettensis Pl. 4, fig. 2 Pl. 4, fig. 4 Pl. 4, fig. 5

K I NHM H DC141 C93796 JHC R. frequens 1203 R. frequens Pl. 8, fig. 1 Pl. 7, fig. 7 L R. frequens J JHC 931 OUM J37756 R. evoluta R. evoluta Pl. 8, fig. 3

strength of the ribbing on the outer phragmocone is so 1935 Ringsteadia bassettensis Spath, p. 41, pl. 12, variable, however, that the name is now reserved for evo- fig. 3a–b. lute forms where the ribbing fades and is lost at diameters 1947a ?Decipia (?Pomerania) marstonensis Arkell, p. 377, of 150–200 mm (Pl. 3, fig. 6; Pl. 4, fig. 3). In forms with pl. 78, fig. 8a, b. bolder, often coarse ribbing at these diameters are 1948 Pachypictonia marstonensis (Arkell); Arkell, p. 402. included in R. bassettensis. The specimens figured by Type material. Holotype, designated by Spath (1935), figured Dohm (1925) and Wilczynski (1962) from north-west Salfeld, 1917, pl. 11, fig. 2a, b, NHM C15426, from the Marston Poland should not be allocated to R. marstonensis. Admit- Ironstone of Wootton Bassett. Var. strattonensis; type of var. tedly, they have the moderately coarse primary ribbing (Taunton Castle Museum 3670) from the Marston Ironstone of characteristic of the marstonensis group, but this is super- Stratton St Margaret, Swindon. imposed on flattened whorl sides running up to a tall, acute, rounded venter. British specimens of R. marstonen- Measurements. Holotype: D(max) 115 mm; at 115 mm, 0.35, sis have an elliptical whorl section (Text-fig. 6A, C). 0.31, 0.40. SMC J44972 (Pl. 3, fig. 5): D(max) 118 mm; at 118 mm, 0.30, 0.25, 0.46.

Ringsteadia bassettensis Spath, 1935 Description. Spath (1935) introduced this name for the more Plate 3, figure 5, Plate 4, figures 1, 2, 4, 5; Text-figures 6B, D–G coarsely ribbed variants of R. marstonensis Salfeld (1917). Bold, rod-like, rectiradiate primary ribs are present at 23 mm, and 1917 Ringsteadia marstonensis Salfeld, pl. 11, fig. 2a–b. these continue with little change to 115 mm (Pl. 3, fig. 5), where 36 PALAEONTOLOGY, VOLUME 53

AB C D TEXT-FIG. 7. Sutures of Ringsteadia SM88 SM290 SM257 SM208 and Pictonia from South Marston and south Dorset. Sutures traced from photographs of the specimens. Numbers 1, 2 and 3 correspond to the three suture types described in the text. A, R. pseudocordata;B,I,R. pseudoyo; C–F, J, R. brandesi; G–H, R. marstonense;L, R. evoluta;M,R. frequens;K,P. baylei; E 2 N, P. densicostata. SM33 3 2

2

2 F SM250

I SM289 H 3 SM39 G SM62 1

1 1

J M SM155 JHC 934 K L NHM C93795 JHC 935 N SMC J47867 2 2 2 2 2

there are 30 rpw. The secondaries arise in a very irregular fash- of Ringsteadia, and fits nicely as a variety of R. bassettensis. The ion, either by bifurcation from the primaries, with one intercala- distinctive features of var. strattonensis (named after Stratton St tory, or by passing of primaries into sheaves of three Margaret, the village on the north-eastern outskirts of Swin- secondaries. Middle whorl fragments have coarse primary rib- don) are the ribbing and whorl section, demonstrated in the bing, approximately 6 per quarter whorl. The ribbing may be two fragments illustrated in Plate 4, figures 1–2, with cross-sec- gently raised (Pl. 4, fig. 5) or markedly raised as in Perisphinctes tions in Text-figure 6D, G. The early ribbing is very low, (Kranaosphinctes) ariprepes Buckman (Pl. 4, fig. 4). The whorl becoming almost nonexistent at the diameter of 70–80 mm section is completely rounded, sometimes broader than high (Pl. 4, fig. 2). In the later part of the phragmocone (Pl. 4, figs (Text-fig. 6B, E–F). 1, 2), there is the sudden development of very coarse, wedge- Var. strattonensis. Arkell (1947a) figured a coarsely ribbed shaped ribs. The whorl breadth is 120 per cent of the whorl aulacostephanid as ?Decipia (?Pomerania) marstonensis, and height (Text-fig. 6D, G). The suture, well seen in Plate 4, figure later ascribed the specimen to Pachypictonia (Arkell, 1948). 2, is clearly of the normal Group 3 Ringsteadia type (see This species is, however, an extremely coarsely ribbed variant below). WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 37

Stratigraphical range. R. bassettensis is moderately common in secondaries (Pl. 6, fig. 4). In some specimens, a smooth, spiral the Pseudoyo and Pseudocordata subzones. band separates secondaries and primaries (Pl. 6, figs 1, 5). This is also seen in Salfeld’s plate 10, figure 1a (incorrectly included Remarks. So close is the style of ribbing of R. bassettensis in R. pseudocordata). to that of P. (Kranaosphinctes) spp. sensu Głowniak (2002) in the mid to late whorls that it may only be pos- Stratigraphical range. R. frequens typifies the Evoluta Subzone as sible to allocate fragmentary specimens to the correct represented by the Osmington Mills Ironstone, the Ringstead taxon, Ringsteadia or Kranaosphinctes, if the suture is visi- Clay in north Dorset and the uppermost Ampthill Clay in Wilt- shire. ble, noting the distinct angle between the first and second lateral lobes of Kranaosphinctes (Wright, 2003, fig. 9b) Remarks. R. frequens is distinguished from R. brandesi by compared with the parallel lobes of evolute Ringsteadia differences in the ribbing. The secondaries in R. brandesi (e.g. Text-fig. 7H). arise in triplicate sheaves or groups of three leading on directly from the primary (Pl. 2, figs 2–3). In R. frequens, a slight forward flexure of the primary rib leads to a Ringsteadia frequens Salfeld, 1917 broadened primary from which arise indistinct sheaves Plate 6, figures 1, 4–5; Plate 7, figures 1, 7; Plate 8, figure 1; of three or four secondaries (Pl. 6, fig. 4; Pl. 8, fig. 1). Text-figures 6H–I, K, 7M Constrictions can be present in both morphospecies, but in R. frequens, these may be followed by a flared 1917 Ringsteadia frequens Salfeld, p. 81, pl. 9, figs 2a–b, 3. primary rib (Pl. 7, fig. 7). The maximum whorl breadth 1917 Ringsteadia brandesi Salfeld, p. 77, pl. 11, fig. 3a–b. in R. frequens is not as close to the umbilicus as in 1917 Ringsteadia pseudo-cordata Blake emend. Salfeld, R. brandesi (Text-fig. 6H, K). The specimen figured as pl. 10, fig. 1a–b. R. frequens by Dohm (1925), although quite close to ?1917 Ringsteadia frequens Salfeld, p. 81. R. frequens, has a cross-section changing from oval- non 1925 Ringsteadia frequens Salfeld; Dohm, p. 28, pl. 5, whorled to fastigate. Salfeld (1917) recorded three speci- fig. 7; pl. 6, fig. 1; pl. 8. fig. 1. mens of R. frequens from the Upper Korallenoolith of the non 1925b Ringsteadia frequens Salfeld; Buckman, pl. 589. Hannover region (Galgenberg, Hoheneggelsen), and one from northern France (Argilles brunes de Cricqueboeuf of Type material. Two syntypes; lectotype (Salfeld, 1917, pl. 9, Villerville, south of Cap de la He`ve). fig. 2a–b, NHM C24092) designated by Buckman (1925b, pl. 589), from the uppermost Ampthill Clay (not Marston Ironstone) of Wootton Bassett. This is the first of the two Ringsteadia evoluta Salfeld, 1917 specimens in Salfeld’s plate, and as such is justifiably regarded as type. Second syntype from ?Osmington Mills Ironstone of Plate 6, figures 3, 6; Plate 8, figures 3–5; Sandsfoot Castle. Text-figures 6J, L, 7L

Measurements. Lectotype: D(max) 130 mm; at 70 mm, 0.46, 1917 Ringsteadia evoluta Salfeld, p. 84, pl. 12, fig. 1a–b. 0.32, 0.31; at 102 mm, 0.39, 0.26, 0.31; at 127 mm, 0.36, 0.24, 1972 ?Decipia sp.; Malinowska, pl. 12, fig. 3. 0.35. DC141 (Pl. 7, fig. 7): D(max) c 97 mm; at 90 mm, 0.40, ?1988 Ringsteadia marstonense Salfeld; Mesezhnikov, 0.27, 0.30. JHC 1203: D(max) 182 mm; at 182 mm; 0.38, 0.20, pl. 9, fig. 13. 0.29. NHM C93796 (Pl. 8, fig. 1): D(max) 113 mm; at 113 mm, 2006 Ringsteadia evoluta Salfeld; Matyja et al., fig. 4d–e. 0.37, 0.25, 0.31. OUM J13104 (Pl. 7, fig. 1): D(max) 138 mm; at 118 mm, 0.38, 0.22, 0.32. SMC J44958 (Pl. 6, fig. 4): D(max) Type material. Holotype (Museum of Geology and Palaeontolo- 101 mm; at 101 mm, 0.42, 0.26, 0.26. 27 ribs. SMC J44959 gy of Go¨ttingen University Collection, 449 ⁄ 1) from the ‘lower (Pl. 6, fig. 5); D(phg-ad) c. 150 mm, D(max) c. 180 mm, Kimmeridge Clay or Upper Corallian of Osmington near 0.25 wh bch, ribs ?25 at 110 mm. SMC J44964: D(phg-?ad) Weymouth’ (Salfeld, 1917, p. 84) – presumably the Osmington 185 mm; at 148 mm; 0.39, 0.27, 0.34. Mills Ironstone of Osmington Mills.

Description. The lectotype is moderately evolute (‘i’, Text- Measurements. BGS GSM5770: D(max) 148 mm; at 148 mm, fig. 8C). The last quarter of a whorl seen is starting to uncoil, a 0.35, 0.22, 0.35. DC122 (Pl. 8, fig. 5): D(phg-ad) 150 mm, feature also seen in Plate 7, figure 1. Before that, the whorl D(max) c. 170 mm, 0.2 wh. bc. At 124 mm, 0.36, 0.26, 0.37. 26 height is greater than the umbilical width. Specimens from south ribs at 80 mm. DC139 (Pl. 6, fig. 3): D(max) 63 mm. At Dorset display the typical strong, prorsiradiate primary ribbing 63 mm; 0.36, 0.29, 0.38. JHC 931 (Pl. 8, fig. 3): D(max) on the inner whorls, ranging from 21 to 29 rpw, with marked 114 mm. At 114 mm, 0.32, 0.23, 0.38. JHC 932 (Pl. 8, fig. 4): constrictions and raised ribs (Pl. 6, fig. 4; Pl. 7, fig. 7). The sec- D(phg-juv) 115 mm, D(max) 130 mm, 0.2 wh bch; at 130 mm, ondaries tend to arise from a broadening of the primary below 0.31, 0.21, 0.43. JHC 934: D(phg-?ad) 175 mm, D(max) the mid-point of the whorl side passing into sheaves of 3 or 4 190 mm, 0.1 wh bch; at 170 mm, 0.37, 0.24, 0.38, 26 ribs at 38 PALAEONTOLOGY, VOLUME 53

D 45 Whorl height %

40

35 x k Pictonia spp. I m 30 o x n Umb.dia. % 25 C Whorl height % 45 R. pseudoyo i x 40

35 R. frequens j

30 R. evoluta + R. marstonensis Umb.dia. % 25 B Whorl height % 45 R. pseudoyo group c x x 40 d f R. brandesi group x 35 x g e x increasingly involute h increasingly evolute 30

A R. marstonensis group Umb.dia. % 25 b a 45 Whorl height %

c 40 R. pseudoyo group d

x f g 35 R. brandesi group e h 30 R. marstonensis group Umb.dia. % 25 10 15 20 25 30 35 40 45 50 55 60

TEXT-FIG. 8. Plots of the ratios of whorl height against umbilical diameter, expressed as a percentage of the shell diameter, in Ringsteadia and Pictonia. A, Pseudoyo Subzone, Marston Ironstone; B, Pseudocordata Subzone, Westbury Ironstone; C, Evoluta Subzone, Osmington Mills Ironstone; D, Baylei Zone, Inconstans Bed, Dorset coast. X = type specimen not from the locality or horizon in question. 8A, shows a good spread through the whole range of morphologies from the highly involute R. pseudoyo group to the evolute R. marstonense group. a, R. pseudoyo (lectotype); b, R. sphenoidea (holotype); c, R. brandesi (lectotype); d, R. anglica (lectotype); e, R. pseudocordata (holotype); f, R. marstonensis (holotype); g, R. bassettensis var. strattonensis (type of var.); h, R. bassettensis (holotype). In 8B, the Pseudocordata Subzone fauna plots differently, with the pseudoyo group very scarce, and a predominance of the evolute marstonense group. 8C shows that the range of variation in the Evoluta Subzone is less than in the earlier subzones. Almost all Ringsteadia can be accommodated within two morphospecies, R. frequens and R. evoluta, although continued use of R. pseudoyo for the more involute forms, and R. marstonensis for depressed forms, is still necessary. (i) R. frequens (lectotype); (j) R. evoluta (holotype). In 8D specimens of Pictonia plot out very differently, all lying at the evolute end of the plot line and within a comparatively close scatter of dots. There is thus a distinct break between the Ringsteadia and Pictonia populations. Unlike Ringsteadia, species in Pictonia are determined by the style of ribbing, not the proportions of the conch. k, P. densicostata var. dorsetensis (type of var.); l, P. seminudatus;m,P. densicostata (holotype); n, P. densicostata var. praebaylei (type of var.); o, P. seminudatus var. ringsteadensis (type of var.). WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 39

170 mm. JHC 935: D(phg-juv) 130 mm, D(max) 160 mm, the Swabian Jura from the tizianiformis Horizon of the 0.5 wh bch; at 145 mm, 0.34, 0.25, 0.41. Hauffinaum Subzone, a horizon that Melendez (2002) has proposed as the highest Oxfordian horizon in the Description. The holotype (Salfeld, 1917, pl. 12) is of an unusual Sub-Mediterranean Upper Oxfordian. smooth, rapidly uncoiling variety. However, it does show one characteristic feature of the species, which is the umbilical seam lying at about the mid-point of the whorl side at medium diam- Genus PICTONIA Bayle, 1878 eters (‘j’, Text-fig. 8C), uncoiling further on the body chamber. Plate 8 figure 5 is more typical of the species, with its irregular, gently raised ribbing. Sheaves of 3 or 4 secondaries may arise Type species. Pictonia baylei Salfeld, 1913 (pro Pictonia cymodoce from a broadening of the primary below the mid-point of the Bayle, 1878 non d’Orbigny, 1850), by subsequent designation by whorl side, but these alternate with frequent strong, simple ribs ICZN under its plenary powers in Opinion 426, 1956 [Bulletin associated with constrictions (Pl. 6, fig. 3; Pl. 8, fig. 3). More on Zoological Nomenclature 14, 259–280]. Bayle’s text to his Atlas coarsely ribbed forms may show bifurcation of secondaries never having been published, the interpretation of his genus (Pl. 6, fig. 6), the bifurcation points being well below the mid- became focussed on his illustrations (Bayle, 1878, pl. 66, figs 1, point of the whorl side. The ribs begin to fade about 70 mm 2, from the Baylei Zone Fm. des Calcaires coquilliers of Cap de ` diameter, although gently raised primary ribs continue to la Heve, Le Havre). These became the type series of Pictonia 150 mm in most specimens, passing indistinctly into very faint baylei, q.v. below. secondaries, which fade over the venter. Remarks. The genus Pictonia comprises medium sized, Stratigraphical range. R. evoluta is by far the most abundant spe- moderately evolute to evolute aulacostephanids that have cies of Ringsteadia in the Evoluta Subzone of both north and sharp, bold primary ribbing and more subdued bifurcat- south Dorset. ing secondary ribbing on the early to middle whorls with frequent additional secondaries intercalated in between Remarks. R. evoluta is best distinguished from R. mar- bifid pairs. Between three and five strong constrictions stonensis by rib style. In R. marstonensis, bold primary ribs are usually, but not always, developed on the middle pass into sheaves of three to six secondaries. In R. evoluta, whorls, each followed by a flared primary rib. The ribbing although in the early whorls there may be strong prima- ends suddenly and the last half whorl of phragmocone ries with biplicate secondaries (Pl. 6, fig. 1; Pl. 7, fig. 3) and body chamber are smooth. The suture is simple as in as in R. marstonensis, the primaries in R. evoluta rapidly some Ringsteadia. In general, Pictonia is an easily recog- become indistinct, with the frequent insertion of constric- nized taxon, the vast majority of specimens showing the tions and simple ribs. Sheaves of three secondaries are distinctive simple suture, constrictions and flared ribs. indistinctly connected to the primaries. The whorl sides Some platycone and depressed varieties lack prominent are flatter in R. evoluta (Text-fig. 6L), distinct from the constrictions, however. round-whorled R. marstonensis. On the later part of the Previous studies of Pictonia have used either ‘vertical’ phragmocone, R. evoluta may have a smooth spiral band (morphological) or ‘horizontal’ (biospecific) classifica- on the whorl side (Pl. 8, fig. 5). tions. Hantzpergue’s major study of Pictonia (Hantzper- The distinction from Pictonia is also clear. Throughout gue, 1989) is almost entirely biospecific. The vast the ribbed portion of its whorl, the strong primaries of majority of Pictonia coming from what Hantzpergue Pictonia bifurcate into secondaries with frequent interca- considered to be one isochronous population in the lated ribs running in between the bifid pairs. In contrast, Lower Kimeridgian of Normandy was allocated to the R. evoluta (and also R. frequens) has only gently raised one biospecies Pictonia baylei Salfeld. This was divided primaries that normally pass below the mid-point of the into Type 1 (with flared ribs) and Type 2 (flared ribs whorl side into sheaves of three or four secondaries. The absent). Each type was subdivided into three groups strong constrictions and flared simple ribs of some ranging from compressed to depressed in whorl section. R. evoluta (Pl. 6, figs 3, 6) affect the ribbing over the ven- Four specimens separated off by Hantzpergue (1989) as ter much more than is usual in Pictonia. P. thurmanni (Contejean) were subsequently allocated to Malinowska’s (1972) ?Decipia sp. from the Pseudocor- Pictonia normandiana Tornquist by Schweigert and Call- data Zone of Czestochowa in Poland matches R. evoluta omon (1997). Earlier forms of Pictonia from the Incon- well, although having slightly denser secondary ribbing. stans Bed of southern England were allocated by The specimen of Ringsteadia figured by Mesezhnikov Hantzpergue (1989) to Pictonia densicostata biosp., again (1988) from the Ural River Basin of Siberia matches clo- subdivided into six groups divided between two main sely some of the Dorset specimens in its gentle primary types. A list of the criteria used by previous authors to ribbing and very faint secondaries. Schweigert and Callo- distinguish the two principle biospecies of Pictonia are mon (1997) recorded R. evoluta from Lochengebiet in given in the Appendix. 40 PALAEONTOLOGY, VOLUME 53

The problem with this system, as was pointed out by Type Material. Holotype (BGS GSM25481) from the Inconstans Sykes & Callomon (1979), is that when, as in this case, Bed of Ringstead Bay, Dorset; var. dorsetensis type of var. there is overlap in morphology between the two biospe- (DK31) from the Inconstans Bed of Black Head; var. praebaylei, cies, often one can only allocate a name (P. baylei biosp., type of var (SMC J47870) from the Inconstans Bed of Ringstead P. densicostata biosp.) to a specimen if one knows which Bay; var. grandicostata, type of var. (SMC J47868) from the In- constans Bed of Ringstead Bay. horizon the specimen came from. Then, the cumbersome reference to, for example, P. baylei, Type 2, Group 3, is Measurements. SMC J47867 (Pl. 9, fig. 2): D(max) 87 mm; at necessary. For this reason, I have preferred to use mor- 82.5 mm, 0.33, 0.27, 0.42; ribs 38 at 15 mm, 39 at 28 mm, 30 at phospecific names in this account, although introducing 60 mm, 31 at 85 mm. OUM J67110 (Pl. 10, fig. 4): D(max) new names only as varieties of pre-existing morphospe- 97 mm: at 97 mm; 0.29, 0.24, 0.49; ribs 35 at 63 mm, 37 cies. Biohorizons can then be distinguished by the varying at 95 mm. proportions of the different morphotypes present in suc- Var. dorsetensis: DK6 (Pl. 7, figs 5–6): D(max) 75 mm, com- cessive populations. pletely septate; at 75 mm, 0.32, 0.30, 0.44; ribs approximately Two main groups of Pictonia are recognized here cen- 23 at 75 mm. DK31. Type of var. (Pl. 9, figs 5–6): D(max) tred around the existing morphospecies P. densicostata 91 mm. At 86 mm; 0.34, –, 0.43, ribs 30 at 58 mm, 29 and P. seminudata. These are comparable with the two at 37 mm. SMC J47912: D(max) c155 mm, 0.6 wh. b.c.; at types of Hantzpergue (1989). The P. densicostata group 94 mm, 0.33, 0.31, 0.40, 26 ribs at 95 mm, 31 at 35 mm, 34 at 18 mm. has marked constrictions and flared ribs, varying from Var. praebaylei: DK5: D(phg-ad) 140 mm+, D(max) 215 mm, depressed (P. densicostata var. dorsetensis nov. and close to aperture; at 200 mm, 0.31, 0.24, 0.48. DK32: D(phg-ad) P. densicostata var. grandicostata nov.) through P. densi- 106 mm, D(max) 124 mm, 0.3 wh bch; at 108 mm, 0.29, 0.23, costata s.s. to P. densicostata var. praebaylei nov. (com- 0.48. NHM C93792: D(max) 98 mm; at 97 mm, 0.29, 0.23, 0.48; pressed). The P. seminudata group lacks marked 27 ribs at 92 mm, 25 at 60 mm. NHM C93793 (Pl. 9, fig. 7): constrictions and flared ribs and is divided into depressed D(phg-ad) 164 mm, D(max) 188 mm, 0.3 wh bch; at 133 mm, forms (P. seminudata) and compressed forms (P. seminu- 0.27, 0.19, 0.49; 33 ribs at 125 mm, 32 at 100 mm, 28 at data var. ringsteadensis nov.). 85 mm, 28 at 67 mm, 31 at 37 mm. SMC. J47444: D(phg-juv) Detailed measurements have been made of rib density, 65 mm, D(max) 85 mm, 0.5 wh bch.; at 84 mm, 0.32, 0.29, but the plots of these unfortunately have little value. 0.48; ribs 29 at 80 mm, 23 at 40 mm. SMC J47870. Type of var. Heavily constricted species of Pictonia may have a marked (Pl. 7, fig. 2): D(phg-ad) 139 mm, D(max) 150 mm, 0.2 wh bch.; at 104 mm, 0.30, 0.25, 0.45; 31 ribs at 100 mm, 27 at fall in rib density between 30 and 70 mm diameter caused 60 mm. SMC J47913 (Pl. 9, fig. 1): D(max) 126 mm; at by a significant proportion of the whorl being occupied 126 mm, 0.31, 0.26, 0.48; ribs 30 at 20 mm, 24 at 45 mm. with constrictions and flared ribs, giving a U-shaped Var. grandicostata: SMC J47868. Type of var. (Pl. 10, fig. 2): curve. Less constricted species may show a straight-line D(phg-ad) 94 mm, D(max) 127 mm, 0.75 wh bch,; at 101 mm, graph. However, changes in rib density can lead to con- 0.30, 0.28, 0.45; ribs 25 at 75 mm, 27 at 85 mm. stricted specimens having a straight-line graph and un- constricted specimens showing a U curve. Thus, rib Description. Up to diameters of 10 mm on the holotype, the pri- curves are not presented here, as they can be misleading. mary ribbing is fine and strongly prorsiradiate. By 15 mm, fine, rectiradiate, biplicate ribbing has developed. A number of sec- ondary ribs extend down almost to the umbilical area at this Pictonia densicostata (Salfeld MS) Buckman, 1924b diameter. The number of primary ribs is 43 at 33 mm, with 37 at 56 mm. Strong constrictions are present, four per whorl, each Plate 7, figures 2, 5–6; Plate 8, figure 6; Plate 9, figures 1–3, followed by a strongly raised primary rib that divides into two 5–8; Plate 10, figures 2, 4; Text-figures 7K, N normally developed secondary ribs. The whorl height is slightly greater than the whorl breadth. 1923 Pictonia densicostata Salfeld; Lamplugh, Kitchin The holotype is more densely costate than is commonly the and Pringle, pl. 2. case in this morphospecies. Most specimens have between 35 1924b Pictonia densicostata Salfeld-Kitchin-Pringle; and 40 rpw at 30 mm, falling to 31 to 36 per whorl at 60 mm. Buckman, pl. 533. In Plate 8, figure 6, at 25 mm, bold primaries bifurcate into ?1927 Pictonia costigera Buckman, pl. 716. prominent secondaries, with deeply intercalated secondaries 1933 Pictonia baylei Salfeld; Arkell, pl. 34, fig. 6. between bifid pairs. The intercalation of secondaries between 1935 Pictonia baylei Salfeld; Spath, pl. 8, fig. 4a–b. bifid pairs continues to 100 mm, when the ribbing suddenly 1947b Pictonia densicostata Salfeld; Arkell, pl. 4, fig. 1. fades. Between three and five marked constrictions per whorl are ?1989 Pictonia baylei Salfeld; Hantzpergue, pls 22a, 24a. present, each followed by a prominently raised primary rib ?1997 Pictonia densicostata Buckman; Matyja and (Pl. 9, fig. 2). In the majority of specimens, the last prominent Wierzbowski, pl. 5, figs 5–11. constriction and raised rib occurs between 60 and 70 mm. Less 2006 Pictonia densicostata Buckman; Matyja et al., commonly, these may continue to 80 mm. fig. 5d. WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 41

Variation. Dorsetensis var. nov. is more coarsely ribbed on the and have marked ventral collars. However, in this aspect, middle phragmocone, with 25–30 rpw. Wide, shallow constric- the Polish specimens are quite close to the densely ribbed tions, 5 per whorl, are each followed by a markedly flared pri- specimen from Wootton Bassett figured as P. baylei by mary rib, whose bifid secondaries run strongly over the venter Spath (1935) and resemble a juvenile specimen figured giving a shallow ventral collar (Pl. 7, figs 5–6). The whorl section herein from the Inconstans Bed (Pl. 8, fig. 6). The range is markedly rounded and depressed in the early whorls. Between of rib density and whorl height ⁄ breadth ratio in var. prae- 70 and 90 mm diameter this changes to a compressed, oval whorl section. The umbilical seam lies at the mid-point of the baylei match those of P. baylei, and Hantzpergue (1989) whorl side with a rapid increase in whorl height and thickness included such specimens within P. baylei biosp. Morpho- and a consequent deep umbilicus (Pl. 9, fig. 5). An excellent logically, however, var. praebaylei is quite distinct, having specimen from Hildenley, North Yorkshire, preserved in a sep- broad, gently rounded ribbing with an oval whorl section, tarian concretion, is in the BGS collection (GSM Y1679). whereas P. baylei has sharp, rod-like ribbing set upon a In praebaylei var. nov., less prominent constrictions are quite smooth rounded-rectangular whorl section. The speci- marked up to a diameter of 70 mm, and there is no marked flar- mens from Zarnglaff (Czarnogłowy), Poland, figured by ing of the subsequent primaries (Pl. 7, fig. 2). Normally, constric- Dohm (1925) as P. baylei, bear stout lateral tubercles and tions do not affect the secondary ribbing, although occasionally should be grouped with Rasenia. there can be some interruption (Pl. 9, fig. 3). Less coarsely ribbed forms of var. praebaylei have become smooth by 110 mm. The ribbing fades out in characteristic irregular manner, faint, simple Pictonia seminudata (Buckman, 1924) ribs running across the almost smooth outer phragmocone. The whorl section throughout is characteristically slim, higher than Plate 5, figures 4, 5; Plate 6, figure 2; Plate 8, figure 2; broad. Half a whorl of almost smooth, septate phragmocone is Plate 10, figures 1, 3 followed by a smooth body chamber. The last septum on NHM C93795 is at about 165 mm, making the maximum diameter ?1896 Pictonia orbignyi Tornquist, 1896, pp. 29–30. 240 mm allowing for a full whorl of body chamber. 1924a Triozites seminudatus Buckman, pl. 494. In most specimens of P. densicostata, the rib style of the mid- 1935 Rasenia orbignyi (Tornquist) (Arkell MS). dle phragmocone continues that of the inner phragmacone, but in some specimens there is a marked strengthening of the pri- Type material. Holotype (BGS GSM47535) from the Inconstans mary ribbing (Pl. 10. fig. 2). Such forms may be styled grandi- Bed of Ringstead Bay. Var. ringsteadensis. Type of var. (DK8) costata var. nov. Both specimens of var. grandicostata figured from the Inconstans Bed of Ringstead Bay. here (Pl. 9, fig. 8; Pl. 10, fig. 2) show that much denser, fainter primary ribbing sets in at 75 mm, with only vestigial secondaries Measurements. OUM J13337: D(max) c. 180 mm, 0.25 + wh. and that the shell becomes completely smooth at 90 mm. bc.; at 110 mm, 0.34, 0.33, 0.43; ribs approximately 28 at 110 mm. BGS GSM25500: D(max) 107 mm; at 63 mm, 0.37, Stratigraphical range. P. densicostata s.s. typifies the densicostata 0.40, 0.49; at 107 mm; 0.34, 0.31, 0.40; 28 ribs at 17 mm, 31 at horizon of Birkelund and Callomon (1985) (see also Matyja 42 mm,. 31 at 107 mm. BGS Zf3723: D(max) 116 mm; At et al. 2006). It is absent in the baylei and normandiana horizons 100 mm, 0.31, 0.33, 0.45; 28 ribs at 45 mm. OUM J67109 (Pl. 5, of Normandy (Hantzpergue 1989). Var. dorsetensis occurs in the fig. 4): D(phg-?ad) 149 mm; at 149 mm, 0.35, 0.33, 0.46. OUM densicostata horizon and possibly in the baylei horizon, there J69569: D(max) 71 mm; at 70 mm, 0.37, 0.41, 0.36; 30 ribs at being a possible match with one Normandy specimen figured by 70 mm. Tornquist (1896, pl. 5, fig. 1), this specimen in Caen Museum, Var. ringsteadensis: DK8. Type of var. (Pl. 10, fig. 1): D(phg- and unfortunately destroyed during the war. Var. praebaylei is juv) 80 mm, D(max) 120 mm, 1 wh bch; at 116 mm; 0.31, 0.23, common in the Inconstans Bed of Ringstead (densicostata hori- 0.42; ribs, 27 at 19 mm, 28 at 34 mm, 31 at 70 mm, 32 at zon). Two specimens figured by Hantzpergue from the baylei 110 mm. OUM J37712 (Pl. 10, fig. 3): D(max) 270 mm, 0.75 wh. horizon (Fm. des Calcaires coquilliers) of Normandy match it bch; at 210 mm, 0.30. –, 0.46; ribs 30 at 150 m, 30 at 105 mm, quite closely. 15 per half whorl at 58 mm, 14 per half whorl at 28 mm. SMC J47448: D(phg-ad) c. 205 mm, D(max) 245 mm, 0.4 wh bch; at Remarks. Pictonia costigera (Buckman, 1927) has dense 180 mm, 0.35, 0.27, 0.43; ribs 35 at 113 mm, 34 at 78 mm. ribbing on the inner whorls similar to that of P. densicos- tata, but the type specimen clearly suffered a major injury Description. Buckman had an unfortunate tendency to make at about 40 mm, so that a deformed quarter of a whorl is holotypes of material that should never have been figured, and succeeded by an untypically boldly ribbed late phragmo- P. seminudata is no exception. One side of the holotype, which is clearly in the matrix of the Inconstans Bed of Ringstead Bay, cone and early body chamber that may be a response to is almost completely abraded away, and on the other, only por- the injury. The specimens from Poland figured as P. dens- tions of mid and outer whorl are visible. A section of ribbed icostata by Matyja and Wierzbowski (1997) match the whorl at about 50 mm diameter shows bold primaries, which holotype in their density of ribbing on the inner whorls trifurcate into secondaries. There is a very gentle ‘S’-shaped but are less regularly constricted than British specimens curve to the ribbing. The last ribbing is visible at 140 mm, and 42 PALAEONTOLOGY, VOLUME 53 the smooth outer phragmocone does not continue the stout, in the early whorls, and the dense secondary ribbing is depressed section of the early whorls, but becomes quite slim (at much less prominent that the primary ribbing. Primaries 155 mm, height 32%, thickness 27%). The maximum diameter and secondaries are equally as bold in P. normandiana. of the specimen is 260 mm, the last septum at 200 mm. A phase P. seminudata and P. normandiana have similar whorl of closing up of sutures is visible at 180 mm, but this is followed sections, but P. normandiana has bolder secondary ribbing. by a quarter of a whorl of normal sutures. The suture is that of Whilst surveying the Sedgwick Museum collection in mainstream Pictonia, with the lateral lobe slightly longer than the suspensive lobe, the development of a second lateral, and 1935, Arkell attached the manuscript name Rasenia orbi- three small elements in the suspensive lobe. gnyi (Tornquist) to the specimen SMC J47448, here Much better material of this oval to round-whorled, involute included in var. ringsteadensis. Arkell’s difficulty was that species is available in the OUM and BGS collections. The umbil- the absence of constrictions and flared primary ribs on the ical seam lying below the midpoint of the whorl side in the early middle whorls of this specimen made allocation to Picto- whorls gives a somewhat cadicone appearance (Pl. 8, fig. 2). Up nia difficult. It is now clear that such forms of Pictonia to 65 mm diameter, the primary ribs are sharp, rod-like, gener- without constrictions and flared ribs are common in the ally rectiradiate, but occasionally with a tendency to develop a Inconstans Bed fauna. These are grouped here in P. semin- gentle ‘S’-shaped swing. No constrictions or flared ribs are visi- udata. Attribution to Rasenia is incorrect. The excellent ble on the inner whorls of most specimens (Pl. 5, fig. 5), but preservation of specimen DK8 (Pl. 10, fig. 1), collected Pl. 8, fig. 2, a juvenile specimen, shows two marked constrictions from the same horizon and locality as SMC J47448, shows per whorl with slightly flared ribs up to a diameter of 30 mm. Where no constrictions are present, there is a distinct tendency that on the innermost whorls of var. ringsteadensis, the to develop them, marked by the presence of regular, fainter, less primary ribs do not show the forward-projecting bullae raised primary ribs every quarter of a whorl, well seen in BGS characteristic of Rasenia. The bold, sharp primary ribbing, GSM25500 and BGS Zf3723. At 65 mm, the primaries become high bifurcation points, the less prominent, bifid second- gently rounded and prorsiradiate (Pl. 5, fig. 5). At 100 mm, the aries and the fading of the secondaries over the venter are prorsiradiate primary ribs join more rectiradiate secondary ribs, all typical of Pictonia seminudata. The dimensions of var. three per primary (Pl. 5, fig. 4), with 12 to 13 primary ribs per ringsteadensis are close to those of the specimen from half whorl. The secondaries run strongly over the venter. Normandy described as P. orbignyi by Tornquist (1896), but Tornquist did not figure his specimen that, being in Variation. Ringsteadensis var. nov. is slimmer and more planu- Caen Museum, was presumably destroyed during the war. late, but again with strong, slightly prorsiradiate primary ribs. Most primaries bifurcate into less prominent secondaries just above the mid-point of the whorl side, with additional secondar- Genus MICROBIPLICES Arkell, 1936 ies intercalated in between bifid pairs, although occasional trifid groups of secondaries are present. The last prominent constric- Type species. Ammonites microbiplex Quenstedt, 1887–88 (p. 876, tion is at 18 mm diameter in Plate 10, figure 1. NHM C77998 pl. 94, fig. 36). Holotype from the Bimammatum Zone of bears its last constriction at 30 mm. In Plate 10, figure 3, the Lochengrundle, Swabian Alb. ribs continue to the end of the phragmocone. This feature is not present in specimen SMC J47448, which has a complete whorl Remarks. Microbiplices comprises small (c. 100 mm) to of smooth phragmocone. Plate 10, figure 1 is almost complete with most of its body chamber preserved and is immature. very small (c. 40 mm) microconch aulacostephanids with coarse, biplicate ribbing and lappets. Secondary ribbing Stratigraphical range. This species is moderately common in the becomes visible below the umbilical seam of the uncoiling Inconstans Bed (densicostata horizon) of Ringstead Bay and body chamber. Quenstedt’s drawing was refigured by occurs sporadically in the Inconstans Bed of Swindon (J.H. Arkell (1936, pl. C, figs 6a–e). One feature of the drawing Callomon Collection). difficult to understand is that the artist has drawn the specimen as completely septate, but in the cross-section, Remarks. Closely allied specimens from Normandy were septae are not shown on the outer whorl. As the outer described as P. thurmanni (Contejean) by Hantzpergue whorl is uncoiling and appears to be body chamber, (1989). Schweigert and Callomon (1997) noted that the use drawing of some of the septae here may have been a mis- of this species in this context was inadvisable, P. thurmanni representation by the artist. being a submediterranean species. The specimens in Arkell (1947a, p. 378) commented that Microbiplices question were referred to P. normandiana Tornquist by had a close affinity with Prorasenia. Schweigert and Callo- Schweigert and Callomon (1997). The differences between mon (1997) proposed that all Pseudocordata and Baylei P. seminudata and P. normandiana sensu Schweigert and Zone aulacostephanid microconchs should be incorpo- Callomon were spelt out by Hantzpergue (1989), noting that rated in the latter. However, it seems best to keep the two P. seminudata is more densely ribbed in the inner whorls, genera separate (Matyja and Wierzbowski 2003). Microbi- and is bigger (250 mm+). P. seminudata is more involute plices is used here for microconchs having biplicate or WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 43 triplicate ribbing on the phragmocone, with bi- or trifur- Remarks. Variants occur with slightly denser ribbing (Pl. 1, cation points fairly high on the whorl side. Prorasenia is fig. 8; Pl. 2, figs 7–8; Sykes and Callomon, 1979, pl. 121, figs used for later forms that have swollen, prorsiradiate pri- 14–15), resembling M. microbiplex (Quenstedt), holotype mary ribs, with trifurcation points on the phragmocone refigured by Arkell (1936, pl. C, figs 6a–c). This has a well below the mid-point of the whorl side, allowing the slightly depressed whorl section on the body chamber, but secondary ribbing to be seen in the umbilicus well before this is rare, and most specimens that have been figured (e.g. the development of the body chamber. Dorn, 1930, pl. 7, figs 4a, b) show a round whorl section. The specimen figured by Brochwicz-Lewinski and Rozak (1975) from Poland belongs to the larger sized group, and Microbiplices anglicus Arkell, 1947 typically has denser ribbing close to the aperture. It came Plate 1, figures 6–8; Plate 2, figures 7–8; Plate 4, figure 6 from the Grossouvrei Subzone (?Caledonica Subzone). Malinowska’s (1972) specimen is quite close to M. anglicus, 1947a Microbiplices anglicus Arkell, p. 348, pl. 76, although lacking the bold, sharp ribbing on the inner figs 6a, b, 7a, b. whorls. Its age is younger than given by Malinowska, ?1972 Perisphinctes (Microbiplices) anglicus Arkell; probably lower Bimmammatum Zone (= Pseudocordata Malinowska, p. 26, pl. 9, fig. 4. Zone) (Wierzbowski, pers. comm. 2007). ?1975 Microbiplices sp. Brochwicz-Lewinski and Rozak,_ pl. 1, fig. 1. 1986 Microbiplices aff. anglicus Arkell; Wright, fig. 6H. 1998 Microbiplices aff. anglicus Arkell; Wright, fig. 7B. Genus PRORASENIA Schindewolf, 1925

Type material. Holotype (BGS GSM30627) from the Sandsfoot Type species. Prorasenia quenstedti Schindewolf, 1925, nom. nov. Grit of Weymouth. for Ammonites biplex bifurcatus Quenstedt, 1887–88, pl. 101, fig. 14, from the White Jura Beta (uppermost Bimmamatum Zone + Planula Zone = Baylei Zone) of Grat bei Laufen in the Swabian Measurements. DC61 (Pl. 1, fig. 8): (D(phg-ad) 24 mm, D(max) Alb. 40 mm, 0.8 wh bch, with terminal constriction, swollen final rib and lappets: at 39 mm, 0.30, 0.30, 0.44; 20 ribs on the last three quarters of a whorl. DC97 (figured, Wright, 1998, fig. 7B): D(phg- Remarks. Schindewolf (1925) did not designate a holo- ?ad) 18 mm, D(max) 28 mm, 0.75 wh bch; at 28 mm, 0.34, –, type for his new species, listing a type series of specimens 0.45. RHUL ic376d: D(phg-?ad) 22 mm, D(max) 34 mm, 0.75 wh including the one figured by Quenstedt and two in his bch. RHUL ic377 (Pl. 4, fig. 6): D(phg-?ad) 20 mm, D(max) own collection. Subsequently, he figured a specimen from 30 mm, 0.75 wh bch; at 29 mm, 0.32, 0.37, 0.52. the White Jura Beta of Beuren in the Swabian Alb as lec- totype (Schindewolf 1926, pl. 19, fig. 1). As defined by Description. When the ribbing is first visible at 12 mm, the Schindewolf (1925, 1926), Prorasenia comprises small primaries are markedly prorsiradiate, bifurcating into rectiradiate (c. 35 mm) microconch aulacostephanids with triplicate secondaries with every third rib trifurcating. At 16 mm, bifurca- Rasenia-like ribbing on the inner whorls. On the outer tion is the norm, the bifurcation points being just visible in the whorl, the ribs become swollen, sharp and biplicate. Sec- umbilicus (Pl. 4, fig. 6). The umbilicus is very deep, with the ondary ribbing is visible in the umbilicus from a very whorl breadth increasing very rapidly, so that at 15 mm, the whorl breadth is twice the whorl height. On the body chamber, early stage. the slightly prorsiradiate primaries average 24 per whorl, varying Use of this taxon for Sub-Boreal and Boreal forms after from 18 to 28 rpw. The slightly less bold secondaries bifurcate Spath (1935) must be provisional. Prorasenia was intro- with extreme regularity, with a smooth band along the venter, duced for forms originating in Sub-Mediterranean areas this band not being present on the last third of a whorl, where where the assumed macroconch counterpart is not at all the ventral ribbing is strong (Pl. 1, fig. 6; Pl. 2, fig. 7). The whorl typical of Sub-Boreal Pictonia. We are probably dealing in section is slightly broader than high. The body chamber uncoils, Britain with microconch aulacostephanids which homeo- clearly revealing the bifurcation points just below the umbilical morph Sub-Mediterranean forms rather than forms that seam (Pl. 4, fig. 6). The average maximum diameter of the 10 have a direct genetic connection. specimens that are complete with the aperture is 49 mm, the holotype being almost complete at 43 mm. Three specimens, DC152, NHM C19475 and NHM C82014, are unusual in having Prorasenia sp. maximum diameters between 60 and 66.5 mm. With more material, it may be possibly to separate off the larger forms as a Plate 1, figure 9; Plate 7, figure 8 separate species. Measurements. NHM C75301 (Pl. 1, fig. 9). D(max) 33.5; Stratigraphical range. M. anglicus is abundant in the Pseudocor- 0.3 wh bc. At 33.5, 0.31, 0.42, 0.45. 17 ribs at 33.5 mm, 15 ribs data Subzone and rare in the underlying Pseudoyo Subzone. at 19 mm. 44 PALAEONTOLOGY, VOLUME 53

Description. Specimen NHM C75301, collected from the charac- Description. SF1 shows the characters of the species very well teristic iron-shot micritic oolite of the Osmington Mills Iron- (Pl. 7, fig. 9). When the primary ribbing is first seen at 8 mm stone of the Fleet Lagoon by Dr N.H. Morris (Pl. 1, fig. 9) diameter, it is strongly prorsiradiate, and trifurcation into sec- shows the characters well. The test is preserved, so the exact ondaries is clearly visible beneath the overlapping whorl. At position of the last septum is not certain, but the specimen is 15 mm, the ribbing becomes only slightly prorsiradiate. The pri- characteristic of a microconch adult with one third of a whorl mary ribs are markedly raised just below the trifurcation points. of body chamber. The primary ribbing is bold, coarse and At 17 mm, the first bifid pairs are present, with no trifurcation strongly raised, prorsiradiate up to 15 mm diameter and then after 23 mm, and less raised primaries. Simple ribs are present rectiradiate. The secondaries trifurcate regularly up to 23 mm, close to the aperture. The whorl height is about equal to the followed by a quarter of a whorl of alternations of biplicate and whorl breadth. The fragmentary DK35 (Pl. 9, fig. 4) has very triplicate secondaries and then biplicate on the body chamber. similar inner whorls but is mature at a larger diameter (cf. On NHM C75302 (Pl. 7, fig. 8), solely biplicate ribbing sets in 42 mm as against 34 mm). earlier, at 17 mm. On NHM C82034, alternations of biplicate and triplicate secondaries are seen to 19 mm, the maximum Stratigraphical range. The holotype of P. bowerbanki (NHM diameter preserved. The secondaries are normally hidden by the 24712) came from Wootton Bassett, Wiltshire, horizon not overlapping whorl, only appearing beneath the uncoiling body recorded. Specimens from Staffin, on the Isle of Skye, show that chamber at 20 mm (Pl. 7, fig. 8). The whorl section is very P. bowerbanki typifies principally the flodigarriensis horizon, depressed in the early whorls, with the whorl breadth twice the coming just above the Oxfordian ⁄ Kimmeridgian boundary whorl height, the whorl section becoming more rounded on the (Matyja et al. 2006, fig. 4), and probably continuing up into the body chamber. densicostata horizon. Thus, at South Ferriby, it is likely that the Oxytoma Cementstone (which yielded the specimen figured in Stratigraphical range. This boldly ribbed, depressed form of Pl. 7, fig. 9) represents the flodigarriensis horizon, slightly older Prorasenia is characteristic only of the Evoluta Subzone. that the densicostata horizon of the Inconstans Bed and that this very distinctive species is a good marker for the base of the Remarks. The specimens discussed here retain the high Kimmeridgian. bifurcation ⁄ trifurcation points of Microbiplices, with sec- ondaries only becoming visible beneath the uncoiling Remarks. The change from the higher bifurcation ⁄ trifur- body chamber. However, they undoutably represent an cation points of Microbiplices to bifurcation ⁄ trifurcation early form of Prorasenia, with swollen ribs on the inner points below the mid-point of the whorl side in P. bower- whorls (similar to P. bowerbanki), and they do not show banki is very striking, the secondaries being visible below the presence of the thin primary ribs characteristic of the umbilical seam at 8 mm diameter. In the specimens Microbiplices. The specimens illustrated as Microbiplices ⁄ figured by Matyja and Wierzbowski (1997), trifurcate sec- Prorasenia transitional form by Matyja et al. (2006) have ondaries become visible in the umbilicus at a much later ribbing characteristic of Microbiplices on the inner whorls stage in the coiling than in the holotype or SF1, suggest- up to 15–20 mm diameter. This feature is not shown in the ing that the Polish specimens are an early variety transi- present specimens, which are thus included in Prorasenia. tional from Microbiplices.

Prorasenia bowerbanki Spath, 1935 Prorasenia cf. hardyi Spath, 1935 Plate 7, figure 9; Plate 9, figure 4 Plate 8, figures 7, 8; Plate 10, figures 5, 6

1935 Prorasenia bowerbanki Spath, p. 43, pl. 14, 1935 Prorasenia hardyi Spath, pl.11, fig. 3a, b, pl. 15, figs 3a, b. fig. 5a, b 1997 Prorasenia bowerbanki Spath; Matyja and 1947b Prorasenia sp. Arkell, pl. 4, fig. 2. Wierzbowski, pl. 5, figs 12, 13. 1978 Prorasenia aff. bathyschista (Koerner); 2006 Prorasenia bowerbanki Spath; Matyja et al., fig. 4j. Wierzbowski, p. 3, fig. 8. 1997 Prorasenia cf. hardyi Spath; Schweigert and Type material. Holotype (NHM 24712) from the ‘Lower Callomon, pl. 7, fig. 10. Kimmeridge Clay’ of Wootton Bassett. 2006 Prorasenia hardyi Spath; Matyja et al., fig. 4i.

Measurements. SF1 (Pl. 7, fig. 9): D(max) 34 mm, complete with Type material. Holotype (NHM 89045) from the ‘Lower Kim- aperture and lappets partially preserved; test partly present meridge Clay’ of Wootton Bassett. obscuring sutures; at 34 mm, 0.30, 0.33, 0.47: ribs 22 at 34 mm, 20 at 17.5 mm, 20 at 14 mm, 24 at 8 mm. DK35 (Pl. 9, fig. 4): Measurements. DK33 (Pl. 10, fig. 6): D(phg-ad) 18 mm, D(max) D(phg-ad) 23 mm+, D(max) c. 42 mm, aperture partially pre- 27 mm, 0.7 wh bch, aperture not present; at 26 mm, 0.33, 0.38, served; at 23 mm, 0.33, 0.34, 0.46; ribs 9 per half whorl at 0.46; 13 ribs per half whorl at 26 mm. DK34 (Pl. 8, fig. 8): 23 mm. D(phg-ad) 22 mm, D(max) 25 mm, 0.3 wh bch; at 20 mm, WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 45

0.38, 0.53, 0.43; 20 ribs at 25 mm, 17 ribs at 11 mm. SMC second lateral lobe is usually set at an angle of some 30 J47351 (Pl. 10, fig. 5): D(phg-ad) 20 mm, D(max) 30 mm, degrees to the first lateral lobe, so that the second lateral 0.75 wh bch, aperture and lappet partly preserved; at 30 mm, saddle widens out in a forwards direction, the two lateral 0.32, 0.37, 0.45; 26 ribs at 30 mm, 26 at 21 mm, 26 at 8 mm. lobes pinching together at its neck. In all aulacostephanids, the second lateral lobe is an exact, miniature copy of the Description. The specimens have the whorl breadth substantially first lateral lobe, set almost exactly parallel to it, so that the greater than the whorl height. The primary ribs are markedly second lateral saddle maintains its width along its length. prorsiradiate in the umbilicus and still prorsiradiate on the body chamber. They are raised where they join the secondaries below This basic pattern is maintained regardless of involute or the mid-point of the whorl side. The secondaries run in a recti- evolute coiling, coarse or fine ribbing, complexity of sutural radiate direction across the venter, frequently interrupted by a development, and comparative length of external, lateral near-smooth band along the centre line. Up to a diameter of and suspensive lobes. Study of the specimens of Ringsteadia 13 mm, each primary trifurcates (Pl. 8, fig. 7). Thereafter, every collected from the Keypoint site, South Marston (Pseudoyo third primary rib trifurcates and the others bifurcate up to a Subzone) shows that they can be arranged into three point a quarter of a whorl into the body chamber, when bifurca- equal-sized groups based on suture: tion is the norm (Pl. 10, fig. 6). The trifurcation points are Group 1 (Text-fig. 7G–I) has a complex suture, with clearly visible in the umbilicus from a diameter of 12 mm an elaborate, deeply incised first lateral lobe and a similar, onwards. slightly smaller second lateral lobe, and an extremely elaborate, backwards projecting suspensive lobe. The Stratigraphical range. The holotype of P. hardyi (NHM 89045) suspensive lobe has six auxiliary elements projecting in an was collected from the Lower Kimmeridge Clay of Wootton Bas- sett (Spath 1935), but the exact horizon is uncertain. Specimens acute, radial direction at about 20 degrees from the radial resembling this species are quite common in the south Dorset line. The second of these six auxiliaries is the longest and Inconstans Bed, typifying the densicostata horizon. On Skye, projects radially almost to meet the second lateral lobe. P. hardyi continues up into the Normandiana Subzone of the Group 2 (Text-fig. 7A, C–E, J) has a moderately to Baylei Zone (Matyja et al. 2006). elaborately complex suture, usually with smaller, less elab- orately developed first and second lateral lobes and a Remarks. The Inconstans Bed specimens have a rib den- backwards projecting suspensive lobe about equal in sity and maximum size matching those of the holotype length to the outer lobes. The suspensive lobe again con- (Spath 1935, pl. 15, fig. 5a, b). However, they are more tains six auxiliaries, moderately or elaborately developed depressed and evolute than the holotype, and may not be and projecting at about 45 degrees to the radial line. cospecific. In addition, the specimens figured in Plate 10 Although the second auxiliary is again the longest, it ends do not show the swollen primary ribs typical of the inner a considerable distance from the second lateral lobe. whorls of P. hardyi. Such thin primary ribbing is nor- Group 3 (Text-fig. 7B, F) has simple sutures, with mally typical of Microbiplices; however, the depressed short, stocky, only moderately incised first and second body chamber and early appearance of secondary ribs in lateral lobes, and a long, radially directed suspensive lobe the umbilicus at only 12 mm diameter are both typical of with six small auxiliaries that are either spirally directed Prorasenia. The Polish specimen figured by Wierzbowski or run at 20 or 30 degrees to the spiral direction. This (1978, pl. 3, fig. 8) has the coarse early ribbing but is feature was described by Arkell (1935, p. xxxii) as the nearly as evolute as the Dorset specimens. The specimen auxiliaries swinging round into line with the external and from southern Germany figured as P. cf. hardyi by lateral lobes to form morphologically not only a ‘second’ Schweigert and Callomon (1997), is more coarsely ribbed but a ‘third’ and ‘fourth lateral’. and more involute than the Dorset specimens and is thus There is no marked tendency for one suture pattern to closer to the holotype. occur preferentially within one particular morphospecies rather than another. The Group 1 style suture is devel- oped throughout R. pseudoyo, R. brandesi, R. pseudo- VARIATION IN THE SUTURE OF cordata and is very common in R. marstonensis and RINGSTEADIA AND PICTONIA R. bassettensis. Group 2 is equally developed throughout the same morphospecies. Group 3 again occurs in From its beginnings as Decipia and Pseudarisphinctes at the R. pseudoyo, R. brandesi and R. pseudocordata, although beginning of the Late Oxfordian, through its substantial there is the possibility that it may be more common in development as Ringsteadia, and into Kimmeridgian Picto- the smooth, evolute varieties of R. marstonensis. This sim- nia and Rasenia, the family Aulacostephanidae maintained ilarity in range of suture pattern throughout all these a characteristic suture pattern that helps separate it from morphospecies is good evidence that we are dealing with the subfamily Perisphinctinae. In the latter subfamily, the one vary variable biospecies. 46 PALAEONTOLOGY, VOLUME 53

Evoluta Subzone Ringsteadia have a much less variable figured from the Unzha River Basin, North-east of suture pattern. The sutures of all specimens I have exam- Moscow, by Rogov and Kiselev (2007). Mesezhnikov ined can be accommodated easily in Group 2 (Text- (1988) has figured Ringsteadia close to R. evoluta from fig. 7L–M). The suture is thus only moderately incised, the Ural River Basin in Siberia. with short lobes, and the second lateral lobe precisely Pictonia and Prorasenia are abundant again at Staffin matching the first in spiral orientation. The auxiliary ele- (Matyja et al. 2006), and at least two species of Pictonia ments project at about 45 degrees to this direction, the reached eastern Greenland (Sykes and Surlyk 1976; suspensive lobe usually being shorter than the lateral or Birkelund and Callomon 1985). However, northwards in external lobes, although in one specimen it is the longest. Svalbard, only Amoeboceras is known (A. Wierzbowski, Ringsteadia is not the only perisphinctoid to show vari- pers. comm. 2003). Eastwards, it seems likely that Pictonia ation in its suture. Glowniak (2002) has reported consid- was unable to progress from northern France into erable variation within the suture patterns of Middle Germany; Early Kimmeridgian strata seem to be present, Oxfordian Kranaosphinctes. Arkell (1939, p. vii) noted but there are no records of these Early Kimmeridgian that the complexity of suture pattern and variation in the ammonites from the area (Arkell 1956, p. 138). However, proportionate length of lobes in Arisphinctes varies con- the range of Pictonia does extend northeastwards to the siderably. Kheta River Basin in Siberia, from where Mesezhnikov Most Pictonia have quite a simple suture (Text-fig. 7K, (1969) figured P. involuta Mesezhnikov and P. ronkinae N). Lobes are of moderate length, rarely deeply incised, Mesezhnikov, both species displaying most of the charac- so that they are usually quite stocky, and the septae teristics of mainstream Pictonia, but being more involute widely separated. The second lateral lobe is usually orien- than western forms. tated parallel to the first, although in occasional speci- mens it may be orientated at 15 degrees or 20 degrees to it, directed slightly radially. The suspensive lobe is direc- Records from the Sub-Mediterranean Province ted obliquely, variable in length, rarely with more than four auxiliaries visible. Slightly more complex sutures are It is clear that Sub-Boreal Ringsteadia made its way at occasionally present (Text-fig. 7K). times into the Sub-Mediterranean Province. R. cf pseu- doyo, R. brandesi and R. pseudocordata (Pseudocordata Subzone) all have been found in the Czestochowa-Krakow RECORDS OF RINGSTEADIA, area of southern Poland, and R. evoluta (Evoluta Subz- PICTONIA,ANDTHEIR one) has been recorded from the Swabian Alps through MICROCONCHS FROM OUTSIDE to southern Poland. However, forms endemic to the Sub- ENGLAND AND NORTHERN FRANCE Mediterranean Province rapidly evolved: Ringsteadia salfeldi Dorn has been recorded by numer- Records from the Boreal and Sub-Boreal Provinces ous authors from the Hypselum Subzone of the Sub- Mediterrannean Province [Brochwicz-Lewinski (1976), Ringsteadia and Microbiplices are abundant at Staffin in Gygy (1995), Schairer (1989), Schweigert and Callomon the Isle of Skye (Matyja et al. 2006). Northwards, there (1997)]. The lectotype (Dorn 1925, pl. 22, fig. 3, desig- do not appear to be any records of Ringsteadia in Green- nated by Gygy 1995) has strong, prorsiradiate, biplicate land or Spitzbergen (Svalbard), further into the Boreal or triplicate, moderately projected ribbing without the Province. However, Sub-Boreal Ringsteadia extended east- sheaves of secondaries characteristic of Sub-Boreal Ring- wards into northern Germany, Salfeld (1917) recording steadia. Strong, rectiradiate simple ribs interrupt the rib- R. brandesi and R. frequens from the upper marly beds of bing every half whorl. R. salfeldi is thus an early form of the ‘Korallenoolith’ of the Hannover region (Text-fig. 1). Ringsteadia, equivalent in age to the Caledonica or Pseu- The specimens of Ringsteadia collected by Dohm (1925) doyo subzone faunas of the Sub-Boreal Province, and from Zarnglaff, now Czarnogłowy in western Poland, developing its own unique features. appear to be a local, endemic group, as they fit precisely Ringsteadia involuta (Quenstedt), as figured by Rozak neither Sub-Boreal nor Sub-Mediterranean species. How- and Brochwicz-Lewinski (1978) from the Hypselum Subz- ever, several species of Ringsteadia and Microbiplices, par- one (Pseudocordata Zone) of Czestochowa in Poland, has ticularly R. brandesi and R. pseudocordata found near the appearance of an evolute R. brandesi, with typical fad- Krakow (Siemiradski Collection, Krakow; taxonomic revi- ing of the ribbing on the mid whorl side, but is more sion by Głowniak and Wierzbowski 2007) have been densely ribbed. recorded from south-central Poland and were listed Ringsteadia submediterranea Wierzbowski, as figured by above. Malinowska (1991) figured Ringsteadia spp. from Wierzbowski (1978, pl. 3, figs 1–3) is probably earliest Central Poland. Ringsteadia close to R. brandesi has been Kimmeridgian in age. Its degree of involution matches WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 47

R. pseudoyo exactly, but the ribbing is much finer and sion marked by the break between ammonite faunas more delicate than anything seen in the Sub-Boreal dominated by the involute Ringsteadia and those domi- province. Vineta streichensis (Oppel) (Schweigert and nated by the evolute, constricted Pictonia (Text-fig. 8C, Callomon 1997, pl. 7, fig. 11) (?early Baylei Zone) is very D). This break was chosen by all the early workers as close to R. submediterranea. defining the Oxfordian ⁄ Kimmeridgian boundary. How- Ringsteadia limosa (Quenstedt), as figured by Wierz- ever, confused by the occurrence of the Early Kimmerid- bowski (1970, 1978), comes from beds equivalent in age gian forms of Ringsteadia listed above, continental to those of the lower Baylei Zone. The cross-section and geologists have generally taken the replacement of the coarse, blunt ribbing match those of R. evoluta well, but ammonite that co-existed with late Ringsteadia, Idoceras the primaries and secondaries are more clearly differenti- planula,bySutneria, at the junction of the Planula and ated, with distinct bifurcation ⁄ trifurcation points. R. limo- Platynota zones, as marking the Oxfordian ⁄ Kimmeridgian sa as interpreted by Schneid (1940, pl. 14, fig. 5), is boundary. Only recently has it become apparent that the rather different, with forward-swept ribbing interrupted boundary has been taken at a substantially higher level by a spiral smooth band on the middle of the whorl side. on the continent than it has been taken in England Ringsteadia flexuoides occurs throughout the Polish (Matyja and Wierzbowski 1994, 1997). A discussion Planula Zone, possibly extending down into the Bimam- of the latest situation as regards correlation of the matum Zone (Wierzbowski 1970). Its stratigraphical Boreal ⁄ SubBoreal and Mediterranean ⁄ SubMediterranean range thus covers the whole Baylei Zone of the Sub- Oxfordian ⁄ Kimmeridgian boundary sequences is given in Boreal scheme. It is more finely ribbed than R. limosa, Matyja et al. (2006). with clear distinction in the early whorls into fine, With the advent of precise absolute dating, sequence rod-like primaries and secondaries with dischizotomous stratigraphy, magnetostratigraphy and chemostratigraphy, separation of the secondaries and forward-swept ribbing. it has become increasingly necessary to define the Oxfor- Schweigert and Callomon (1997) recorded R. flexuoides dian ⁄ Kimmeridgian stage boundary at one synchronous from the Bimammatum Zone of the Swabian Alb level throughout Europe (Wierzbowski et al. 2006). The (probably uppermost Oxfordian). preferred option of these authors is to take it at the base Four species of Pictonia are known, none at all close to of the Baylei Zone, and this is the option taken in the Sub-Boreal forms. P. (Pictonia) densicostata (Matyja and definitive papers of Gradstein and Ogg (2004) and Grad- Wierzbowski, 1997, pl. 5, figs 5–11) is much less regularly stein et al. 2004. Callomon (2004) has argued strongly constricted than is typical in the forms from England fig- that the Kimmeridgian GSSP should be situated in south ured here but is still closely related. Pictonia kuiaviensis Dorset. However, in the author’s opinion, none of the Matyja and Wierzbowski (2002, pl. 1 figs 4–11) is a youn- Dorset exposures recommended by Callomon (2004) as ger form transitional between Pictonia and Rasenia. Picto- suitable sites for a GSSP are satisfactory. The fossiliferous nia perisphinctoides (Wegele) and P. praeperisphinctoides shelly limestone of the Inconstans Bed of Ringstead Bay, Schweigert and Callomon figured by Schweigert and Call- which yielded the vast majority of the specimens of Picto- omon (1997) from the Sub-Mediterranean Province are nia described here, is now covered by sea defences and much less heavily constricted and flared than mainstream beach shingle. All other Inconstans Bed exposures in Sub-Boreal forms. south Dorset are poor and yield only sporadic ammo- nites. In addition, a succession such as that in south Dor- set containing a substantial non-sequence, and in places CONCLUSIONS an unconformity, beneath the earliest Kimmeridgian, although excellent for defining a sequence boundary, is All sections of Upper Oxfordian strata in southern entirely unsuited for defining a stage boundary. Walsh England show a gradual, often step by step, deepening of et al. (2004, p. 209) note that ‘unconformities have long facies, from shallow water sands and ooidal ironstones been rejected as suitable horizons for golden spikes’ (i.e. into deeper water clays and marls. At many localities in GSSP’s) because correlative evidence is only available on England, Oxfordian ⁄ Kimmeridgian boundary strata are one side of the spikes. The proposal of Matyja et al. missing, an erosion surface separating Early Kimmerid- (2006) is to draw the Oxfordian ⁄ Kimmeridgian boundary gian strata from Late Oxfordian strata. The major trans- beneath the flodigarriensis horizon of the Baylei Zone at gression km1 of the Early Kimmeridgian (Taylor et al. Staffin in Skye in a succession that has continuous sedi- 2001) saw the temporary establishment of shallow, mentation across the Rosenkrantzi ⁄ Bauhini and Pseudo- muddy, clastic-dominated seas with an abundant shallow cordata ⁄ Baylei zone boundaries. Further work will be water bivalve ⁄ brachiopod fauna (Inconstans Bed) necessary to decide precisely where this level lies within throughout shelf areas of England. The effect of this the Sub-Mediterranean Bimammatum Zone (A. Wierz- regression ⁄ transgression was to leave a gap in the succes- bowski, pers. comm. 2006). 48 PALAEONTOLOGY, VOLUME 53

Acknowledgements. Professor J.H. Callomon kindly loaned his BIRKELUND, T. and CALLOMON, J. H. 1985. The Kimme- collection of uppermost Oxfordian ammonites, gave much valu- ridgian ammonite faunas of Milne Land, central east Green- able advice and considerably improved the manuscript in his ref- land. Bulletin Grønlands Geologiske Undersøgelse, 153, 56 pp. ereeing of it. Dr B.M. Cox and Professor A. Wierzbowski also BLAKE, J. F. and HUDLESTON, W. H. 1877. On the Coral- made substantial improvements during their refereeing. Mr C. lian rocks of England. Quarterly Journal of the Geological Soci- Griffiths loaned his collection of Oxfordian ⁄ Kimmeridgian amm- ety of London, 33, 260–405. onites. Thanks are due to the respective curators at the Natural BRISTOW, C. R., COX, B. M., WOODS, M. A., PRUD- History Museum London, British Geological Survey, Oxford Uni- DEN, H. C., SOLE, D., EDMUNDS, M. and CALLO- versity Museum, Sedgwick Museum Cambridge and Devizes MON, J. H. 1991. The geology of the A303 trunk road Museum for allowing the author to study and photograph amm- between Wincanton, Somerset and Mere, Wiltshire. Proceed- onites in their care. Thanks are also due to the staff at Mowlem ings of the Dorset Natural History and Archaeological Society, Ltd for allowing access to the Keypoint site near Swindon, and 113, 139–143. for making available specimens collected by workers there. Mr J. BROCHWICZ-LEWINSKI, W. 1976. Oxfordian of the McGrow assisted in the preparation of the figures. Czestochowa Area. I. Biostratigraphy. Bulletin de l’Acade´mie Polonaise des Sciences, Se´rie des Sciences de la Terre, 24, 37–46. Editor. Elizabeth Harper —— and ROZAK, Z. 1975. Time changes of Oxfordian ammonite fauna of the Polish Jura Chain; some reflections. Bulletin de l’Acade´mie Polonaise des Sciences, Se´rie des Sciences REFERENCES de la Terre, 22, 113–125. BROOKFIELD, M. E. 1978. The lithostratigraphy of the upper Oxfordian and lower Kimmeridgian Beds of south ARKELL, W. J. 1933. The Jurassic system in Great Britain. Dorset, England. Proceedings of the Geologists’ Association, Clarendon Press, Oxford, viii+681 pp. 89, 1–32. ——1935. A monograph of the ammonites of the English Coral- BUCKMAN, S. S. 1919–30. Type Ammonites. Published by the lian Beds, part 1. Palaeontographical Society Monograph, 88, author. London and Thame. i–xxxii, 1–30. ——1921. Type Ammonites Part 27, pp. 41–42. ——1936. A monograph of the ammonites of the English Coral- ——1924a. Type Ammonites Part 46, pp. 21–28. lian Beds, part 2. Palaeontographical Society Monograph, 89, ——1924b. Type Ammonites Part 49, pp. 41–48. xxxiii–xlvi, 31–42. ——1925a. Type Ammonites Part 51, pp. 57–64. —— 1937. Report on ammonites collected at Long Stanton, ——1925b. Type Ammonites Part 53, pls 528, 577–592. Cambs., and the age of the Ampthill Clay. Summary of Pro- ——1926. Type Ammonites Part 60, pp. 29–36. gress of the Geological Survey of Great Britain and the Museum ——1927. Type Ammonites Part 63, pls 135, 157, 706–717. of Practical Geology for the year, 1935, 2, 64–88. CALLOMON, J. H. 1963. Sexual dimorphism in Jurassic ——1939. A monograph of the ammonites of the English Coral- ammonites. Transactions of the Leicester Literary and Philo- lian Beds. part 5. Palaeontographical Society Monograph, 92, sophical Society, 57, 36 pp. lv–lxiv, 105–190. —— 1969. Dimorphism in Jurassic Ammonites. 111–125. In ——1940. A monograph of the ammonites of the English Coral- WESTERMANN, G. E. G. (ed.). Sexual dimorphism in fossil lian Beds, part 6. Palaeontographical Society Monograph, 94, Metazoa and taxonomic implications. E. Schweitzerbart’sche lxv–lxxii, 191–216. Verlagsbuchhandlung, Stuttgart, iv+250 pp. —— 1947a. A monograph of the ammonites of the English Cor- —— 1981. Dimorphism in ammonites. 257–273. In HOUSE, allian Beds, part 13. Palaeontographical Society Monograph, M. R. and SENIOR, J. R. (eds). The Ammonoidea: the evolu- 101, 351–378. tion, classification, mode of life and geological usefulness of a —— 1947b. The Geology of the Country around Weymouth, Swa- major fossil group. Systematics Association, London, Special nage, Corfe and Lulworth. Memoir of the Geological Survey of Volume, 18, xvi+593 pp. Great Britain, England and Wales, 386 pp. —— 1985. The evolution of the Jurassic ammonite family Car- ——1948. A monograph of the ammonites of the English Coral- dioceratidae. Special Papers in Palaeontology, 33, 49–90. lian Beds, part 14. Palaeontographical Society Monograph, 102, —— 2004. Some comments on the proposals for the GSSP of i–v, 379–420. the Kimmeridgian Stage. International Subcommission on —— 1956. Jurassic geology of the world. Oliver & Boyd, Edin- Jurassic Stratigraphy, Newsletter, 31, 21–24. burgh and London, xiv+806 pp. CONTEJEAN, C. 1859. E´ tude de l’e´tage kimme´ridien dans les —— KUMMEL, B. and WRIGHT, C. W. 1957. Mesozoic environs de Montbe´liard. Me´moires de la Socie´te´ d’histoire nat- Ammonoidea. L80–L490. In MOORE, R. C. (ed.). Treatise urelle du Doubs, 3rd Series, 4, 352 pp. on invertebrate Palaeontology, Part L, Mollusca 4, Cephalo- CONTINI, D. and HANTZPERGUE, P. 1973. Le Kimme´- poda – Ammonoidea. Geological Society of America, Boulder, ridgien de la re´gion de Montbe´liard. Annales de Science de Colorado and University of Kansas Press, Kansas, xxii+L1– l’Universite´ de Besanc¸on, 18, 143–179. L490. COPE, J. C. W. 1995. Introduction to the British Jurassic. 1–7. BAYLE, E. 1878. Fossiles principaux des terraines. Service de la In TAYLOR, P. D. (ed.). Field geology of the British Jurassic. Carte ge´ologique de´taille´e. Explication de la Carte ge´ologique Geological Society, London, 286 pp. de´taille´e, 4, Atlas, part 1, 158 pls. WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 49

DOHM, B. 1925. U¨ ber den oberen Jura von Zarnglaff i.P. und (northern Poland) and their correlation value. Acta Geologica seine Ammoniten-fauna. Abhandlung Geologisches-Pala¨ontol- Polonica, 52, 411–421. gisches Institut, Universita¨t Greifswald, 2, 1–40. —— —— 2003. Correlation chart of standard chronostrati- DORN, P. 1925. Das Auftreten der Gattung ‘‘Ringsteadia- graphic ammonite zonations at the Oxfordian ⁄ Kimmeridgian Salfeld’’ im unteren Malm der no¨rdlichen Frankenalp. boundary. International Subcommission on Jurassic Stratigra- Zeitschrift der Deutschen Geologischen Gesellshaft, 77, 529–524. phy, Newsletter, 30, 25–27. —— 1930. Die Ammoniten-Fauna des untersten Malm der Frank- —— —— and WRIGHT, J. K. 2006. The Sub-Boreal ⁄ Boreal enalb. 1, Die Perisphincten. Palaeontographica, 73, 107–175. ammonite succession at the Oxfordian ⁄ Kimmeridgian bound- GALLOIS, R. W. 2005. On the Kimmeridgian (Jurassic) suc- ary at Flodigarry, Staffin Bay (Isle of Skye), Scotland. Transac- cession of the Normandy coast, northern France. Proceedings tions of the Royal Society of Edinburgh: Earth Sciences, 96 (for of the Geologists’ Association, 116, 33–43. 2005), 387–405. GŁOWNIAK, E. 2002. The ammonites of the family Peri- MELENDEZ, G. 2002. Oxfordian Working Group Report. sphinctidae from the Plicatilis Zone (lower Middle Oxfordian) International Subcommission on Jurassic Stratigraphy, News- of the Polish Jura Chain (central Poland); their taxonomy, letter, 29, 11–15. phylogeny and biostratigraphy. Acta Geologica Polonica, 52, MESEZHNIKOV, M. S. 1969. Kimmeridgian ammonites. 99– 307–364. 118. In SACHS, W. N. (ed.). The reference section of the —— and WIERZBOWSKI, A. 2007. Taxonomic revision of Upper Jurassic deposits in the Kheta River Basin. Leningrad, the perisphinctid ammonites of the Upper Jurassic (Plicatilis to Nauka, 208 pp. [In Russian]. Planula zones) described by Josef Siemiradzki (1891) from the —— 1988. Oxfordian. 39–45. In KRYMHOLTS, G. Ya., Krakow Upland. Volumina Jurassica, 5, 31–137. MESEZHNIKOV, M. S. and WESTERMANN, G. E. G. GRADSTEIN, F. M. and OGG, J. G. 2004. Geologic Time (eds). The Jurassic ammonite zones of the Soviet Union. Geolog- Scale 2004 – why, how and where next? Lethaia, 37, 175–181. ical Society of America, Special Paper, 223, 116 pp. —— —— SMITH, A. G., BLEEKER, W. and LOURANS, MORRIS, N. J. 1968. Palaeontological and stratigraphical stud- L. J. 2004. A new Geologic Time Scale, with special reference ies in the Upper Jurassic rocks. Unpublished Dr. Phil. thesis, to Precambrian and Neogene. Episodes, 27, 83–100. University of Oxford, 234 pp. GYGY, R. A. 1995. Datierung von Seichtwassersedimenten des MORTON, N. and HUDSON, J. D. 1995. Field guide to the Spa¨ten Jura in der Nordwestschweiz mit Ammoniten. Eclogae Jurassic of the Isles of Raasay and Skye, Inner Hebrides, NW geologicae Helveticae, 88, 1–58. Scotland. 209–280. In TAYLOR, P. D. (ed.). Field geology of HANTZPERGUE, P. 1989. Les ammonites Kimme´ridgiennes the British Jurassic. Geological Society, London, 286 pp. du Haut-Fond d’Europe occidentale – biochronologie, syste´ma- D’ORBIGNY, A. 1842–51. Pale´ontologie franc¸aise. Terrains tique, e´volution, pale´obioge´ographie. Cahiers de Pale´ontologie jurassiques. I. Ce´phalopodes. Victor Masson, Paris, 623 pp. Centre National de la Recherche Scientifique, Paris, 1–427. QUENSTEDT, F. A. 1887–88. Die Ammoniten des schwa¨bis- KUTEK, J. 1968. The Kimmeridgian and uppermost Oxfordian chen Jura III: Der Weisse Jura. E. Schweitzerbart’sche Verlags- in the S.W. margins of the Holy Cross Mountains (Central buchhandlung, Stuttgart, 818–1140. Poland). Part 1, Stratigraphy. Acta Geologica Polonica, 18, ROGOV, M. A. and KISELEV, D. N. 2007. The Kimmerid- 493–574. gian of Russia and adjacent areas, its subdivision and correla- MAKOWSKI, H. 1963. Problem of sexual dimorphism in tion. Geological Institute of RAS, Moscow, 1–31. ammonites. Palaeontologica Polonica, 12, 1–92. ROZAK,_ Z. and BROCHWICZ-LEWINSKI, W. 1978. MALINOWSKA, L. 1972. Middle and Upper Oxfordian in Upper Oxfordian of Czestochowa; some new data. Bulletin de the north-west part of the Czestochowa Jurassic. Instytut Geo- l’Acade´mie Polonaise des Sciences, Se´rie des Sciences de la Terre, logiczny, Biuletyn, 233, 4–67. 26, 47–51. —— 1991. Boreal faunal influences in Upper Oxfordian in north SALFELD, H. 1913. Certain Upper Jurassic strata of England. and central Poland. Prace Panstowowego Instytutu Geologicznego, Quarterly Journal of the Geological Society of London, 69, 423– 135, 1–27. 432. MATYJA, B. A. and WIERZBOWSKI, A. 1994. On correla- —— 1914. Die Gliederung des oberen Jura in Nordwesteuropa. tion of Submediterranean and Boreal ammonite zonations of Neues Jahrbuch fu¨r Mineralogie, Beil Band B, 37, 125–246. the Middle and Upper Oxfordian: new data from central —— 1917. Monographie der Gattung Ringsteadia (gen. nov.). Poland. Geobios, 17, 351–358. Palaeontographica, 62, 69–84. —— —— 1995. Biogeographic differentiation of the Oxfordian SCHAIRER, G. 1989. Ammoniten aus dem Braunen und and Early Kimmeridgian ammonite faunas of Europe, and its Weissen Jura von Sengenthal. Mittelungen der Bayerische stratigraphic consequences. Acta Geologica Polonica, 45, 1–8. Staatssammlung fu¨r Pala¨ontologie und Historische Geologie, —— —— 1997. The quest for a unified Oxfordian ⁄ Kimmerid- 29, 109–131. gian boundary: implications of the ammonite succession at SCHINDEWOLF, O. H. 1925. Entwurf einer Systematik der the turn of the Bimmammatum and Planula zones in the Perisphincten. Neues Jahrbuch fu¨r Mineralogie, Geologie und Wielun Upland, central Poland. Acta Geologica Polonica, 47, Palaeontologie, Beilagerband B, 52, 309–343. 77–105. —— 1926. Zur Systematik der Perisphincten. Neues Jahrbuch fu¨r —— —— 2002. Boreal and Subboreal ammonites in the Sub- Mineralogie, Geologie und Paleontologie, Beilagerband B, 55, mediterranean uppermost Oxfordian in the Bielawy section 495–517. 50 PALAEONTOLOGY, VOLUME 53

SCHNEID, T. 1940. U¨ ber Raseniiden, Ringsteadiiden und Pic- —— 1978. Ammonites and stratigraphy of the Upper Oxfordian toniiden des no¨rdlichen Frankenjura, Abteilung III. Palaeonto- of the Wielen Upland, central Poland. Acta Geologica Polonica, graphica A, 91, 78–119. 28, 299–377. SCHWEIGERT, G. and CALLOMON, J. H. 1997. Der —— COE, A. L., HOUNSLOW, M. W., MATYJA, B. A., bauhini-Faunenhorizont und seine Bedeutung fu¨r die Korrela- OGG, J. G., PAGE, K. N., WIERZBOWSKI, H. and tion zwischen tethyalem und subborealem Oberjura. Stuttgar- WRIGHT, J. K. 2006. A potential stratotype for the Oxfor- ter Beitra¨ge zur Naturkunde Serie B (Geologie und dian-Kimmeridgian boundary: Staffin Bay, Isle of Skye, U.K. Pala¨ontologie), 247, 69 pp. Volumina Jurassica, 4, 17–33. SIEMIRADZKI, J. 1891. Fauna kopalna warstw oksfordzkich i WILCZYNSKI, A. 1962. La stratigraphie du Jurassique supe´ri- Kimerydzkich w okregu Krakowskim i przyległych czesciach eur a` Czarngłowy et Swietoszewo. Acta Geologica Polonica, 14, Kro´lestwa Polskiego. Pamietnik Akademii Umietjetnosci w 3–112. Krakowie Wydzial Matematyczno-Przyrodniczy, 18, 1–92. WILLIAMS, M. E. 2003. The development of hiatal surfaces in SPATH, L. F. 1924. On the Blake Collection of ammonites the Osmington Mills Ironstone Member of the Upper Jurassic from Kachh, India. Palaeontologica Indica, New Series, 9, 1–29. Ringstead Formation of south Dorset, England. Proceedings of ——1935. The Upper Jurassic invertebrate faunas of Cape Leslie, the Geologists’ Association, 114, 193–210. Milne Land, East Greenland, I. Meddelelser om Grønland, 99, WRIGHT, J. K. 1980. The Oxfordian Stage. 61–76. In COPE, 1–80. J. C. W. (ed.). A correlation of the Jurassic rocks of the British STEINMANN, G. 1890. In STEINMANN, G. and DOE- Isles, Part 2, Middle and Upper Jurassic. Geological Society, DERLEIN, L. Elemente der Pala¨ontologie. Leipzig, 848 pp. London, Special Report, 15, 109 pp. SYKES, R. M. and SURLYK, F. 1976. A revised ammonite —— 1986. A new look at the stratigraphy, sedimentology and zonation of the Boreal Oxfordian and its application in north- ammonite fauna of the Corallian Group (Oxfordian) of east Greenland. Lethaia, 9, 421–436. south Dorset. Proceedings of the Geologists’ Association, 97, —— and CALLOMON, J. H. 1979. The Amoeboceras zonation 1–21. of the Boreal Upper Oxfordian. Palaeontology, 22, 839–903. —— 1996. Perisphinctid ammonites of the Upper Calcareous SYLVESTER-BRADLEY, P. C. 1956. The new palaeontology. Grit (Upper Oxfordian) of North Yorkshire. Palaeontology, 39, 1–8. In SYLVESTER-BRADLEY, P. C. (ed.). The species 433–469. concept in palaeontology. Systematics Association Publication, —— 1998. New stratigraphical and palaeoecological data from London, 2, vi+145 pp. the Sandsfoot Formation (Upper Oxfordian) of south Dorset. TALBOT, M. R. 1974. Ironstones in the Upper Oxfordian of Proceedings of the Geologists’ Association, 109, 15–24. southern England. Sedimentology, 21, 433–450. —— 2003. New exposures of the Ampthill Clay near Swindon, TAYLOR, S. P., SELLWOOD, B. W., GALLOIS, R. W. and Wiltshire, and their significance within the sequence of the CHAMBERS, M. H. 2001. A sequence stratigraphy of the Oxfordian ⁄ Kimmeridgian boundary beds in England. Proceed- Kimmeridgian and Bolonian stages (Late Jurassic): Wessex – ings of the Geologists’ Association, 114, 97–121. Weald Basin, southern England. Journal of the Geological —— —— 2001. British Upper Jurassic Stratigraphy (Oxfordian to Society, London, 158, 179–192. Kimmeridgian). Geological Conservation Review Series No. TORNQUIST, A. 1896. Die degenerierten Perisphinctiden des 21, Joint Nature Conservancy Commission, Peterborough, Kimmeridge von Le Havre. Me´moire de la Societe´ pale´ontologi- 266 pp. que Suisse, 23, 1–44. ZIEGLER, P. A. 1982. Post Hercynian plate reorganization in WALSH, S. L., GRADSTEIN, F. M. and OGG, J. G. 2004. the Tethys and Artic-North Atlantic domains. 711–755. In History, philosophy and application of the Global Stratotype IVIANSPEIZER, W. (ed.). Triassic–Jurassic rifting. Conti- Section and Point (GSSP). Lethaia, 37, 201–218. nental breakup and the origin of the Atlantic Ocean and passive WIERZBOWSKI, A. 1970. Some Upper Jurassic ammonites margins. Developments in Geotectonics, 22(B). Elsevier, of the genus Ringsteadia, Salfeld 1913, from central Poland. Amsterdam, 755 pp. Acta Geologica Polonica, 20, 270–285.

APPENDIX Inconstans Bed Pictonia are indeed commonly regularly ribbed (Pl. 9, fig. 2; Pl. 10, fig. 4). However, irregularly 1. Critical review of the main features used to distinguish ribbed forms (Pl. 7, fig. 2; Pl. 9, fig. 5) and forms with between P. densicostata biosp. from the Inconstans Bed of Dorset flexible, slightly curved ribbing (Pl. 5, fig. 4; Pl. 9, fig. 1) and P. baylei biosp. from the ‘Calcaires coquilliers’ of the Nor- are also common. mandy coast. 2. P. baylei biosp. becomes smooth earlier than P. densicostata Hantzpergue (1989) and Schweigert and Callomon (1997) biosp. have proposed the following criteria to distinguish the two bio- In fact, the reverse is true. Twenty-two specimens of species: P. densicostata biosp. that I have measured become smooth 1. P. densicostata biosp. generally has more inflexible ribs, reg- at an average of 102 mm diameter. Hantzpergue’s (1989) ularly disposed. photographs show that P. baylei biosp. becomes smooth at WRIGHT: AULACOSTEPHANIDAE OF OXFORDIAN ⁄ KIMMERIDGIAN BOUNDARY BEDS 51

an average of 108 mm (eight specimens). These measure- C88665d, C89059, OUM J14616, SM25, 44, 53, 58, 192, 227, ments are so close that the point of initiation of smooth 228, 251, 252, 254, 262, 275, 277, 289, 290, 291). All specimens phragmacone must be a very minor distinguishing feature prefixed SM collected from the Keypoint site. One specimen of the two biospecies. (CAMSM X 50142) from the Westbury Ironstone of Westbury. 3. The stage of approximated ribbing is not confined to the Two specimens (OUM J21186, J37527) from the Osmington nucleus in P. densicostata biosp., whilst it is in P. baylei Mills Ironstone (Ringstead Coral Bed) of Ringstead Bay, and biosp. one specimen (BGS Zm2295) from the Osmington Mills Iron- Specimens only rarely preserve the nucleus, and this is stone of the East Fleet section. not a viable criterion to separate most specimens of Pictonia. R. sphenoidea: Holotype (NHM C41691) from the Marston Iron- 4. The ribbing of P. densicostata biosp. divides without atten- stone of Wootton Bassett. Seven specimens (SM43, 46, 84, 96, uation of the secondaries. In P. baylei biosp., the secondar- 146, 269, 270) from the Marston Ironstone at the Keypoint site, ies fade immediately ventral of the point of division, and South Marston. strengthen on the venter. Hantzpergue (1989) does not state at what diameter this R. brandesi: Three specimens from the Marston Ironstone of criterion is present. Prior to 55–60 mm diameter, no fading Wootton Bassett (NHM 15425 (paratype), C36972, C4122695). of the ribbing is visible in P. densicostata biosp. However, Fifty one specimens (NHM C50750, 50770 (lectotype), SM19, between 60 mm and the end of the ribbing, fading of the 21, 26, 28, 33, 35, 38, 45, 47, 51, 52, 73, 81, 83, 90, 95, 113, 118, ribbing ventral to bifurcation points and on the venter is 121, 122, 123, 124, 127, 132, 138, 142, 151, 153, 154, 155, 180, present in almost all specimens (e.g. Pl. 10, fig. 4). Use of 182, 185, 208, 229, 231, 244, 245, 246, 247, 257, 260, 261, 263, this criterion to separate the two biospecies is thus difficult. 266, 272, 274, 292, YORYM:2004.179) from the Marston Iron- 5. The cross-section of P. densicostata biosp. is more oval. stone of South Marston. Specimens prefixed SM collected at the The cross-section of Inconstans Bed Pictonia at c. Keypoint site. Five specimens from the Westbury Ironstone (D 50 mm diameter in fact varies from extremely depressed, 1702, GSM7717, BGS GSM97553, BGS HBW2644, NHM nearly twice as thick as high, to compressed, with height C26457). Five specimens from Unit III of the Sandsfoot Grit, greater than thickness. However, the distinctive, flat-sided four from Sandsfoot Castle (DC58, DC94, ic105, SMC J47616) forms resembling the holotype of P. baylei are not present and one from the East Fleet section (DC146). in the Inconstans Bed, P. densicostata biosp. consistently having the rounded whorl side only present in some R. anglica: five specimens (BGS GSM25502 (lectotype), P. baylei biosp. GSM97551, NHM C41696, SMC J44981, UR 10895) from the 6. P. densicostata biosp. has denser ribbing. Westbury Ironstone. NHM C50768 (paratype) from the Ampt- The data are:- hill Clay (?Marston Ironstone) of Shrivenham and NHM C322 P. baylei biosp. (from Hantzpergue, 1989): 24 specimens, (paratype) from ‘Wiltshire’. Seven specimens (NHM C4864, average 26.7 rpw between 70–80 mm diameter. SM42, 72, 78, 88, 135, 250) from the Marston Ironstone of P. densicostata biosp.: 14 specimens, average 30.2 rpw South Marston. Specimens prefixed SM collected at the Keypoint between 70–80 mm diameter. This is by far the clearest site. distinction, although several specimens from Dorset fall R. pseudocordata: 13 specimens from the Westbury Ironstone below the Normandy average. (BGS 2649Hbr, GSM8254, GSM46264 (holotype), GSM97554, 7. Ribbing of P. densicostata biosp. is finer, sharper and D1704, OUM J21235, J21236, J21237, J21238, J21239, J21240, divides at two thirds of the whorl height. P. baylei biosp. SMC J44980, J44984). 45 specimens from the Marston Ironstone has rounder, coarser ribbing dividing at the middle of the of South Marston (OUM J14615, SM16, 23, 27, 32, 41, 49, 55, whorl side. 56, 72, 74, 75, 79, 114, 116, 117, 119, 120, 128, 145, 183, 188, The ribbing of Pictonia from the Inconstans Bed in fact 189, 195, 209, 230, 232, 249, 253, 259, 264, 265, 287, 288, 293). varies from fine, dense and sharp to blunt and gently All specimens prefixed SM collected from the Keypoint site. One rounded. However, almost all specimens have the bifurca- specimen (NHM C41695) from the Marston Ironstone of Wootton tion point situated at three fifths (not two thirds) of the Bassett. One specimen from the Sandsfoot Grit of Sandsfoot Castle whorl height. The height of division of the secondaries in (DC94). P. baylei biosp. can vary – most specimens do have the division point in the middle of the whorl side, although in R. pseudocordata var. nudus: 10 specimens from the Marston some specimens (OUM JZ 701; Hantzpergue, 1989, pl. 22a) Ironstone of the Keypoint site (SM31, 48, 57, 76, 85, 187, 191, division is at three fifths of the whorl height. 206, 244, 276). R. marstonensis: 38 specimens (NHM C15427 (holotype), C71124, C88665c, C88665f, SM20, 22, 24, 29, 30, 34, 37, 39, 50, 2. List of specimens of Ringsteadia and Microbiplices examined 59, 61, 62, 64, 65, 66, 68, 71, 77, 82, 89, 115, 133, 134, 136, 137, 147, 148, 150, 179, 201, 255, 258. 273, 285) from the Marston R. pseudoyo: five specimens (BGS GSM72900, NHM 24089 (lec- Ironstone of South Marston. Specimens prefixed SM collected at totype), C88651, C88657, OUM J14822) from the Marston Iron- the Keypoint site. Eight specimens from the Westbury Ironstone stone of Wootton Bassett. 21 specimens from the Marston of Westbury (BGS GSM97555, NHM C26451, C88650, OUM Ironstone of South Marston (NHM 36960 (paratype), C71125, J16734, RHUL ic972, SMC J44971, J44983, J44987), one speci- 52 PALAEONTOLOGY, VOLUME 53

men (DC119) from the Sandsfoot Grit of Sandsfoot Castle, P. densicostata var. praebaylei: Type of var. (SMC J47870) and 37 Weymouth, and one specimen (DC126) from the Osmington further specimens (DK5, BGS 1666HBW, BGS GSM25485, Mills Ironstone (Ringstead Coral Bed) of Ringstead Bay. NHM C93791, C93795, OUM J14674, J14675, J14676, J14677, J14679, J14682, J14688, J14689, J14690, J14691, J14694, J14695, R. bassettensis: Holotype (NHM C15426) from the Marston Iron- J14696, J14921, J17831, J20063, J21181, J37713, J37714, J37717, stone of Wootton Bassett. Thirteen specimens (SM40, 63, 67, 69, J37718, J67138, SMC J47444, J77858, J47859, J47860, J47861, 129, 143, 144, 149, 193, 210, 256, 284, 286) from the Marston J47862, J47863, J47871, J47873, J47913) from the Inconstans Bed Ironstone of the Keypoint site, South Marston. Two specimens of Ringstead Bay. One specimen (SMC J47861) from the Nana from the Westbury Ironstone of Westbury (RHUL ic973, SMC Bed of Ringstead Bay. J44972). P. densicostata var. grandicostata: Type of var. (SMC J47868) and R. bassettensis var. strattonensis: Three specimens (SM60, 190, three further specimens (DK32, NHM C93792, OUM J14678) 248) from the Marston Ironstone at the Keypoint site, Swindon, from the Inconstans Bed of Ringstead Bay. and one specimen from the Marston Ironstone of this general area (NHM C71123). One specimen (DC142) from the Sands- P. seminudata s.s.: Holotype (BGS GSM47535) in the matrix of foot Grit, Unit III, of the East Fleet section. the Inconstans Bed of Ringstead Bay, Dorset. 11 topotypes (BGS GSM3018, GSM25500, Zf3723, Zf3726, NHM C3316, C23885, R. frequens: One specimen (NHM C24092, lectotype) from the OUM J13337, J21218, J29499, 67109, J69569). uppermost Ampthill Clay of Wootton Bassett. 20 specimens (DC62, 141, BGS Zf3727, JHC 1203, NHM C3344, C12073, P. seminudata var. ringsteadensis: Type of var. (DK8) and eight C23881, C82419, C93796, OUM J13104, J13478, J19456, SMC further specimens (DK4, 11, NHM C77998, C82407, C93793, J13479, J44958, J44959, J44960, J44964, J44965, J44966, J44967) OUM J20063, J37712, SMC J47448) from the Inconstans Bed of from the Osmington Mills Ironstone of Osmington, Ringstead Ringstead Bay. Bay, Sandsfoot Castle and the East Fleet section. One specimen Microbiplices anglicus: 22 specimens (BGS GSM30627 (holotype), from the Ampthill Clay of the Bourton Bypass. DC55, 56, 61, 63, RHUL ic376a, ic376b, ic376d, ic376e, ic377, R. evoluta: 27 specimens (DC122, 125, 139, BGS GSM5770, NHM C19475, C69773, C82013, C82014, C82015, C82017, GSM3864, Zf3725, JHC 931, 932, 934, 935, 936, NHM C3306, C82018, C82019, C82023, C82025, C82026, C82427) from the C14664, C19467, C82416, C82417, C82418, C93794, OUM Sandsfoot Grit of Sandsfoot Castle, and one specimen each from J14662, J14666, J14700, J14923, J21175, J37756, SMC J44963, the Sandsfoot Grit of the East Fleet (DC97), Black Head (DC61) J44969, J44973) from the Osmington Mills Ironstone of Ring- and Ringstead Bay (DC140). One specimen from the Marston stead Bay, Osmington and the Fleet Lagoon. Twenty speci- Ironstone (SM163) and four specimens (SM297, 298, 299, 300) mens from the Ampthill Clay of the Bourton Bypass, north from the overlying Ampthill Clay of the Keypoint site, south Dorset. Marston; three specimens (NHM 26461 (figd. Arkell, 1947a, pl. 76, figs 1a,b), 25367a, 25367b) from the Westbury Ironstone. P. densicostata s.s.: Holotype (BGS GSM25481) from the Incon- stans Bed of Ringstead Bay, Dorset. Sixteen further specimens Microbiplices triplicata: Three specimens from the Osmington (DK9, GSM Zf3724, OUM J14672, J14683, J14686, J14687, Mills Ironstone of the Fleet Lagoon (NHM C75301, C75302, J14693, J14697, J14924, J17831, J21219, J37716, J44693, J67110, C82034). SMC J47864, J47867) from the Inconstans Bed of Ringstead Prorasenia bowerbanki: One specimen (SF1) from the basal Kim- Bay. meridgian Oxytoma Cementstone of South Ferriby Pit, Lincoln- P. densicostata var. dorsetensis: Type of var. (DK31) from the In- shire, and one specimen (DK35) from the basal Kimmeridge constans Bed of Black Head, and eight specimens (DK6, 8, 24, Clay of the Bourton Bypass, Dorset. GSM 82917, OUM J14919, J21217, SMC J47869, J47912) from Prorasenia cf. hardyi: Four well-preserved microconchs from the the Inconstans Bed of Ringstead Bay, three from Wootton Bas- Inconstans Bed of Ringstead Bay. (SMC J47351, DK10, 33, 34). sett (GSM Y2904, OUM J67111, SMC J29068), and one from the Kimmeridge Clay of North Yorkshire (GSM Y1679).